the iot based environmental sensing platform

The IoT Based Environmental

Sensing Platform

by

Yansong Wang

B.Eng.

A thesis submitted in accordance with the requirements for the award of

the degree of Master of Philosophy of the University of Liverpool

June 2019

P a g e | ii

Declaration

I hereby declare that except where specific reference is made to the work of others, the

contents of this thesis are original and have not been submitted in whole or in part for

consideration for any other degree or qualification in this, or any other University. This

dissertation is the result of my own work and includes nothing which is the outcome of

work done in collaboration, except where specifically indicated in the text.

The copyright of this thesis rests with the author. Copies (by any means) either in full,

or of extracts, may not be made without prior written consent from the author.

Copyright © 2019 Yansong Wang. All rights reserved.

P a g e | iii

I would like to dedicate this thesis to my family and friends

P a g e | iv

Acknowledgements

Looking back to the two years of unforgeable and hard-working MPhil studies, I am

grateful to many people and many things. First and foremost, I would like to express

my sincere gratitude and thanks to my main supervisor Professor Yi Huang, who

provided me the precious opportunity to test myself at the highest level. You have

encouraged and helped me to grow up from a fresh graduate student to a young

researcher. I learned so much from you both academically and personally. Thank you

for the invaluable comments and advice on my research as well as my life and career.

It is a great honor for me to be one of your students, and I hope I have repaid the faith

you showed in me.

I would also like to thank my parents. They have always supported me with no

expectation of a reward. Their continuous help and understanding have made my life

full of love and I am grateful for everything you have done. I would also like to express

my appreciation to my girlfriend. I would never have succeeded without her love,

tolerance, support, encouragement and patience. This work is dedicated to you.

Special thanks are also paid to my brilliant and lovely colleagues and friends; in

particular to Dr Jiafeng Zhou, Dr. Chaoyun Song, Dr. Zhouxiang Fei, Dr. Yuan Zhuang,

Dr. Anqi Chen, Dr. Manoj Stanley, Dr Dajun Lei, Dr Zhenghua Tang as well as

Tianyuan Jia, Wenzhang Zhang, Chen Xu, Qiang Hua, Jingyuan Jiang, Lyuwei Chen,

Jinyao Zhang joseph Sumin, Ahmed Aliedin and Umniyyah Ulfa for many fruitful

discussions and enjoyable moments.

P a g e | v

Table of Contents

Contents

The IoT Based Environmental Sensing Platform ...................................................................... i

Table of Contents ......................................................................................................................... v

List of Figures ............................................................................................................................. vii

Acronyms .......................................................................................................................................ix

Abstract ........................................................................................................................................ xii

Chapter 1 Introduction .......................................................................................................... 1

1.1 Background: Internet of Things .................................................................................................1

1.2 IoT Applications: ...........................................................................................................................6

1.3 Motivation of this work: ........................................................................................................... 11

1.3.1 Motivations...................................................................................................................... 11

1.3.2 Challenges ....................................................................................................................... 12

1.3.3 Objectives ........................................................................................................................ 13

1.3.4 Original Contributions: ................................................................................................. 14

1.4 Thesis Outline: ............................................................................................................................ 15

References: ......................................................................................................................................... 16

Chapter 2 Literature Review ............................................................................................... 22

2.1 Review of Communication Protocols for IoT Applications .............................................. 22

2.1.1 Short-Range Communication Protocols .................................................................. 22

2.1.2 Low Power Wide Area Network (LPWAN) Protocols ............................................ 24

2.2 Review of Wireless Sensor Network ...................................................................................... 29

2.3 Review of Environmental Monitoring Activities ................................................................. 32

2.4 Summary ...................................................................................................................................... 35

References: ......................................................................................................................................... 36

Chapter 3 A Compact Anemometer with Ultra-low Energy Consumption............ 43

3.1 Introduction ................................................................................................................................ 43

3.2 Model Establishment and Theory .......................................................................................... 47

3.2.1 Sensor Design and Measuring Method ................................................................... 47

3.2.2 Model Establishment .................................................................................................... 50

3.3 Experimental Results and Discussions .................................................................................. 54

3.4 Summary ...................................................................................................................................... 60

References .......................................................................................................................................... 61

Chapter 4 An Environmental Sensing Platform with Self-powered Standalone

Weather Stations ....................................................................................................................... 65

4.1 Introduction .......................................................................................................................... 65

4.2 Hardware and Software ..................................................................................................... 67

4.2.1 Hardware ......................................................................................................................... 67

4.2.2 Software ........................................................................................................................... 73

P a g e | vi

4.3 Measurement and Discussion .......................................................................................... 76

4.3.1 Deployment .................................................................................................................... 76

4.3.2 Measurement Results ................................................................................................... 77

4.3.3 Coverage Analysis ......................................................................................................... 81

4.3.4 Energy Performance Analysis ..................................................................................... 83

4.3.5 Discussions ...................................................................................................................... 86

4.4 Summary ............................................................................................................................... 88

4.5 Reference: .................................................................................................................................... 88

Chapter 5 Conclusions and Future Work ........................................................................ 93

5.1 Summary ...................................................................................................................................... 93

5.2 Key Contributions and Limitations ........................................................................................ 94

5.3 Future Work .......................................................................................................................... 96

P a g e | vii

List of Figures

Fig. 1- 1The global IoT market growth [9].................................................................................... 1

Fig. 1- 2 The IoT architecture [2]. (a) Three-layer. (b) Five-layer. ......................................... 4

Fig. 1- 3 IoT application areas [17].................................................................................................. 6

Fig. 2- 1 Three classes of LoRaWAN end devices [26]. .......................................................... 27

Fig. 2- 2 The LoRaWAN architecture [27]. .................................................................................. 28

Fig. 2- 3 Five layers protocol stacks for wireless sensor network [29]. .............................. 29

Fig. 2- 4 A weather station mounted on the adjustable tripod [44]. ................................. 33

Fig. 3- 1 A cup-type anemometer. ............................................................................................... 44

Fig. 3- 2 An ultrasonic anemometer [11]. ................................................................................... 45

Fig. 3- 3 A hot-wire anemometer [16].. ...................................................................................... 45

Fig. 3- 4 The accelerometer, sensor case, and mounting structure.. ................................. 48

Fig. 3- 5 The 3D force analysis of the anemometer.. .............................................................. 52

Fig. 3- 6 The setup of the wind speed measurement. ............................................................ 54

Fig. 3- 7 The measured accelerations of two axes without data processing. ................. 55

Fig. 3- 8 The measured accelerations of two axes with data processing. ........................ 56

Fig. 3- 9 The comparison between the measured wind velocity and the actual wind

velocity. .......................................................................................................................................... 56

Fig. 3- 10 The comparison between the proposed anemometer and the

commercial cup-type anemometer. .................................................................................... 57

Fig. 3- 11 The wind speed measurements of the commerical anemometer and the

proposed anemometer. ........................................................................................................... 58

Fig. 3- 12 The wind direction measurements of the commerical anemometer and

the proposed anemometer..................................................................................................... 58

Fig.4- 1 The architecture of the weather station. ..................................................................... 67

Fig.4- 2 RN2483 LoRaWAN communication module. ............................................................ 70

Fig.4- 3 The inside look of the sensor node. ............................................................................. 70

Fig.4- 4 The outside look of the sensor node. .......................................................................... 72

Fig.4- 5 The LoRaWAN gateway. ................................................................................................... 72

Fig.4- 6 Software working diagram .............................................................................................. 74

Fig.4- 7 The architecture of the environmental sensing platform. ..................................... 75

Fig.4- 8 The deployment of sensor nodes and the LoRaWAN gateway.. ........................ 77

Fig.4- 9 The measured temperature over 5 days. .................................................................... 77

Fig.4- 10 The measured air pressure over 5 days. ................................................................... 78

Fig.4- 11 The measured humidity over 5 days.......................................................................... 78

Fig.4- 12 The measured loudness over 5 days. ........................................................................ 78

Fig.4- 13 The measured air quality over 5 days........................................................................ 79

Fig.4- 14 The measured results of three sensor nodes. ......................................................... 80

Fig.4- 15 The coverage test of LoRaWAN. ................................................................................. 80

Fig.4- 16 The coverage distance of the LoRaWAN gateway. ............................................... 82

P a g e | viii

Fig.4- 17 The current consumption of the sensor node. ....................................................... 84

Fig.4- 18 The change of the voltage level in 3 months.......................................................... 84

Fig.4- 19 The change of the voltage level in 1 month. .......................................................... 85

P a g e | iv

Acronyms

ABP Activating By Personalization

ABS Acrylonitrile Butadiene Styrene

ADR Adaptive Data Rate

API Application Programming Interface

BLE Bluetooth Low Energy

CR Code Rate

DBPSK Differential Encoding Binary Phase-Shift

Keying

FSK Frequency Shift Keying

GFSK Gaussian Frequency Shift Keying

GMSK Gaussian Minimum Shift Keying

GSM Global System for Mobile

HVAC Heating Ventilation and Air Conditioning

ICT Information and Communications

Technology

IMU Inertial Measurement Unit

P a g e | x

IoT Internet of Things

ISM Industrial, Scientific, and Medical

ISO International Organization for

Standardization

ISP In-System Programming

LPP Lowe Power Payload

LPWAN Low-Power Wide-Area Network

LTE Long-Term Evolution

MAC Media Access Control

MCU Microcontroller Unit

NB-IoT Narrowband Internet of Things

NFC Near-Field Communication

OSI Open Systems Interconnection

OTAA Over The Air Activation

PC Polycarbonate

PHY Physical

PMU Power Management Unit

PZT Piezoelectric Transducer

P a g e | xi

RFID Radio-Frequency Identification

RSSI Received Signal Strength Indicator

SC-FDMA Single Carrier Frequency Division

Multiple Access

SF Spreading Factor

SoC System on Chip

SRAM Static Random-Access Memory

TPH Temperature Pressure Humidity

TTN The Things Network

VOC Volatile Organic Compounds

Wi-Fi Wireless Fidelity

WSN Wireless Sensor Network

P a g e | xii

Abstract

Environmental monitoring defined as using sensors to detect the condition of the

ambient environment has attracted an upsurge of research interests. It can be applied in

many fields such as weather prediction, outdoor asset monitoring, structural health

monitoring, etc. IoT technologies act as an essential tool to realize the wireless

connection between weather stations and the real-time data collection from

environmental sensing nodes. One of the ultimate objectives of environmental

monitoring is to deploy sensing devices over a large area to obtain enough data samples

automatically for condition analysis and management. However, current weather

stations are constrained by their high energy consumption, bulky size, limited

accessibility, and high cost, which impedes the wide deployment of environmental

sensing devices. The purpose of this thesis is to develop a new environmental sensing

platform by adopting IoT technologies which could overcome many challenging design

problems in this field. There are two main contributions of this thesis.

The first contribution is the development of an ultra-low energy consumption

anemometer. There are different ways to measure the wind speed, using such as

ultrasonic sensors and hot-film resistors. However, the energy consumptions of those

anemometers are relatively high, and extra power lines need to be connected to power

the anemometer. A novel method to measure the wind speed and direction is developed

which uses a 6-axis accelerometer to measure the change of the acceleration induced

by the wind force and then to calculate the corresponding wind speed. The power

consumption of the proposed anemometer is only 3.42 mW, which is significantly

smaller than other anemometers, thus batteries can be used as the energy source to

power the proposed anemometer instead of the power line. The new design can make

an anemometer to be a standalone device powered by a battery only, which increases

the deployment flexibility of the sensor node. The proposed anemometer has low

energy consumption and compact size which also allows it to be easily integrated to be

P a g e | xiii

part of an environmental sensing platform.

The second contribution of this work is the development of an IoT based environmental

sensing platform which consists of wireless sensor nodes, a communication network,

data visualization and storage cloud. The environmental sensing platform has the ability

of collect, send, present, and store various environmental parameters and it has the

advantages of easy deployment, real-time accessibility, low deployment and

maintenance cost. The sensing device in the environmental platform is a true standalone

device which has wireless connection and can be powered by using a solar panel. The

data gathered by the sensor node can be transmitted to a data visualization platform

named Cayenne in real time, and the user can easily access the collected data from

sensors deployed in different areas.

This thesis has introduced and successfully demonstrated a number of novel methods

for environmental monitoring applications. An environmental sensing platform has

been developed with improved performance in terms of such as reducing the energy

consumption of the device, increasing the deployment flexibility of the end device, and

improving the connectivity of the sensor node. The proposed environmental sensing

platform is of great importance to future research and industry applications.

Chapter 1: Introduction

P a g e | 1

Chapter 1 Introduction

1.1 Background: Internet of Things

The Internet of Things (IoT) is a concept first introduced by Kevin Ashton at Procter &

Gamble (P&G) in 1999 who viewed Radio-Frequency Identification (RFID) as an

essential technology to the Internet of Things [1]. With the rapid development of

advanced communication technologies in the last two decades, different short-range

and long-range communication protocols such as RFID, Wi-Fi, Bluetooth, Bluetooth

Low Energy (BLE), Zigbee, LoRaWAN, and Narrow Band Internet of Things (NB-IoT)

have been widely used to connect a vast number of devices. Nowadays, The Internet of

Things has become one of the hottest emerging technologies which could

fundamentally change the way how people interact with things. IoT can be defined as

things and sensors being connected to the Internet to achieve the interchange of data

and message to realize smart control and management [2]. The Machine-to-Machine

(M2M) traffic volume in the U.S. has a 250 % growth in 2011, and 45 % of the total

internet traffic flows will be constituted by M2M communications by 2022 [3-7]. IoT

has become national strategies of some countries[8].

Fig. 1- 1 The global IoT market growth [9].


P a g e | 2

Different technologies such as information and communications technology (ICT),

wireless sensor network (WSN), radio-frequency identification (RFID), system on chip

(SoC), machine learning, big data, and cloud computing contribute to the development

of IoT. Some latest communication methods like LoRaWAN and NB-IoT were specially

designed for machine-type communication with low energy consumption, wide

coverage, and strong robustness against noise. It provides reliable channels for things

to be connected to the Internet. The wireless sensor network (WSN) is a large number

of sensors connected to form a network. The sensors can provide useful information

about the ambient environment or the conditions of the monitored devices. RFID is a

foundational technology for IoT, which is widely used in logistics, retailing, and supply

chain to identify, track and monitor objects attached with RFID tags. To realize smart

control and management, we need to process and analyze the gathered data by using

machine learning algorithms.

IoT architecture can generally be viewed as a centralized architecture, where a

heterogeneous and dense set of devices deployed over a wide area generate a large

amount of data delivered by using suitable communication technologies to a control

center [10]. Then, the data will be stored and processed for further decision making,

smart control, and other purposes. A key characteristic of an IoT architecture is its

capability of integrating different devices, systems, and technologies into the same

existed centralized platform to support the progressive evolution of the IoT. Also, for a

centralized network, the data should be transferred by using a reliable channel to reduce

the latency and packet errors to provide reliable information in time. There are other

design criteria that need to be concerned, such as security, real-time interaction,

interoperability, and energy consumption to improve the overall performance of the

centralized platform.

Since it is very challenging to design a centralized platform consisting of millions of

devices deployed over a wide range, and some researchers propose new decentralized


P a g e | 3

IoT architectures based on blockchain or fog computing. By adopting the blockchain

technology, it provides the access control policies to improve the mobility, accessibility,

scalability, security, and the transparency of IoT platforms [11]. Instead of transmitting

the data to the cloud to be further processed, the fog computing allows to do the

calculation and make decisions in local to avoid generating a huge volume of data to be

transmitted and reduce the latency [12].

The International Standards Organization (ISO) developed the Open Systems

Interconnection (OSI) model to divide network communication into seven layers which

are the physical layer, the data link layer, the network layer, the transport layer, the

session layer, the presentation layer, and the application layer [13]. The IoT model is

very similar to the communication model, and the basic model is a 3-layer architecture

consisting of perception, network, and application layers [14]. Recently, more models

have been proposed to add more functions to the IoT architecture, and a five-layer

model has been summarized to deliver the full functionality of the IoT, which are the

objects layer, the object abstraction layer, the service management layer, the application

layer and the business layer [2]. IoT architecture can be explained in Fig. 1-2.


P a g e | 4

Fig. 1- 2 The IoT architecture [2]. (a) Three-layer. (b) Five-layer.

A. Objects Layer

The first layer can be viewed as the perception layer, where different sensors and

actuators are used together to sense data from the ambient environment and perform

operations. RFID tags can also be used to track and monitor things. It works as the

fundamental layer of the IoT to gather physical information like weight, motions,

vibrations, temperature, humidity, brightness, positions, and acceleration to be further

processed and analyzed. It provides data to be transferred in the object abstraction layer.

B. Object Abstraction Layer

In this layer, it allows collected data from the objects layer to be transmitted and shared

with other devices or central cloud to realize the interconnections of different things.

The object abstraction layer provides channels for the data to be transferred by using

some advanced communication technologies such as Bluetooth Low Energy (BLE),


P a g e | 5

LTE-M, Wi-Fi, NB-IoT, LoRaWAN, and Zigbee. The communication channel should

be designed to be secure, with low energy consumption, and low latency to unleash the

performance of the IoT networks.

C. Service Management Layer

The role of the service management layer is to provide functionalities to integrate

services and applications in IoT by using middleware technology. It enables IoT

application programmers to work with different objects without considering specific

hardware or software platforms. A well-designed service layer can identify standard

application requirements and provide Application Programming Interface (API) and

protocols to support the required services, applications, and user needs [15]. The data

obtained in the objects layer will be stored and analyzed in this layer by using some

data analysis techniques and data processing algorithms [16].

D. Application Layer

The application layer acts as the interface between the IoT systems and the users. This

layer presents the collected information such as temperature, position, and other data to

the user to make decisions and realize smart control and management. There are some

challenges in this layer that need to be solved to improve the interactions between the

IoT systems and the users. Since there will be numerous devices made by different

manufacturers to be integrated, the cooperation between different devices, platforms,

and clouds is difficult to achieve. The application layer also provides the interface to

the business layer, where high-level analysis and reports can be produced [2].

E. Business Layer

The business layer is on the top of the previous four layers, and it manages the overall

IoT activities and services. The role of the business layer is to build a business model

to provide essential services and to generate profits from the service being provided. It

acts as the driving force of the whole IoT activities and services.


P a g e | 6

1.2 IoT Applications:

The IoT has vast application areas, including smart transportation, smart grid, smart

home, logistics, retailing, environmental monitoring, etc. With the help of IoT

technologies, the traditional manufacturing processes can be fundamentally changed,

and the production efficiency will be dramatically increased. Also, the IoT can be

applied in the home, such as the smart lighting system, smart metering, advanced

heating, ventilation, and air conditioning system (HVAC), etc. It can improve the

comfort level at home and save energy at the same time. The IoT system can generate

a large amount of data, and a lot of useful information which could be obtained from

the database. For example, a wearable blood pressure meter connected to the Internet

can collect and store the blood pressure of the patient over a period, and doctors can

then monitor patients in real time and give appropriate medical advices and treatments

based on the collected data.

Fig. 1- 3 IoT application areas [17].


P a g e | 7

1) Smart Home

An increasing number of home appliances like refrigerator, oven, and washing machine

are now becoming computing-enabled and can be connected to the internet by using

Wi-Fi or Bluetooth. Those Internet-connected devices in the home can provide to the

users with more reliable information about their current status, and the user can have

more convenient control of those home appliances. There are more advantages an IoT

system can bring in a smart home. The smart lighting system and the HVAC system can

offer a comfortable living space and save energy at the same time; some security

cameras, magnetic door, and window alarms, motion sensors, and smoke sensors can

guarantee a safe living space. Some smart speakers are now acting as the interfaces

between the users and home appliances, and users can talk with the smart speaker to

control some home appliances like light bulbs. However, there are some issues raised

by IoT technologies as well, and privacy and data security are vital concerns [18]. With

more devices in the home connected to the Internet, more secure, sophisticated, and

robust data and access protection systems should be designed to avoid cybersecurity

attack.

2) Elderly People Care

The life expectancy of people has continuously increased due to the advanced medical

treatments, and the percentage of the elderly people in total population keeps raising as

well [3, 19]. Some of the elderly people have to live alone, and they will be difficult to

take care of themselves, especially when emergencies happen. A fall-detection system

is vital for aged people to send alarming signals to the corresponding people to offer

help when a fall is detected [20]. The IoT technologies can be a very promising solution

to partially solve the elderly people care problem by deploying some Internet-connected

device to realize real-time monitoring and emergency warning. The design and

deployment of the IoT based elderly people care system are very challenging. Because

elderly people can be relatively difficult to learn new technologies, and the deployed


P a g e | 8

system should be noninvasive so that the new technologies will not change the living

style of the elderly people. Also, the collected data should also be kept safe and secure

to protect the privacy of the user. The upfront hardware investment of the system and

the total cost of the elderly people care service are still too high to achieve a large-scale

deployment [21]. A low cost, noninvasive, reliable and user-friendly elderly people care

system should be designed, and a replicable business model of the elderly people care

service should be developed as well.

3) Health Care

With the help of the wireless sensors and advanced communication technologies, IoT

is redesigning modern health with promising technological, economic, and social

prospects [22]. IoT based health care can be applied to glucose level sensing, oxygen

saturation monitoring, rehabilitation system, medication management, and wheelchair

management. RFID tags can be used to monitor and track the production and

distribution of medicine to guarantee the quality of each medicine. Different wearable

sensors like blood pressure sensors, body temperature sensors, electrocardiograms

sensors, and electromyography sensors can be used to collect vital data to support early

diagnosis, real-time monitoring, and medical emergencies. With the dramatical expends

of the application of wireless sensing technologies in medical treatment, a confined

hospital environment can be extended to the patient’s home to provide a quiet, familiar

and less stressful place for the patent [23]. Some start-up and large companies are

actively involved in the building of IoT medical cloud and database to transmit, store,

and present collected data. A Chinese firm has developed an all-in-one medical imaging

and information management platform supporting cloud-based image storage and

computation, 3-D image post-processing, and visualization [22]. More research and

development efforts should be put in this area to make health care more convenient,

affordable, and effective.


P a g e | 9

4) Logistics

The development of IoT technologies has provided vital supports to establish a new

logistics system to allow the tracking and monitoring of goods on a global scale. The

IoT based logistics system dramatically improves the monitoring efficiency in the

distribution, delivery, storage, and sales of the products, and it can be a useful tool to

build a fast, low-cost, and efficient supply chain. The robots play an essential role in

the smart logistic system, and it can pick, transport, and deliver the products to the right

conveyor by using the robot arms. The RFID tags have the function to track, monitor

and manage the parcels, the RFID sensing doors can be used in congestion with RFID

tags to confirm the types and quantities of products to be delivered [24]. The automatic

driving car can further reduce the cost and improve the safety level of the transportation

of the goods. The adoption of the IoT technologies in logistics can transfer the logistics

industry from a traditional labor-oriented industry to a technology-oriented industry.

5) Smart Cities

With the rapid growth of the global economy, people are more concentratedly living

and working in big cities. This will bring a negative impact to the cities such as the

short of energy sources, polluted air and water, and the traffic congestions, etc. The IoT

technologies can be widely applied in different areas in cities such as waste

management, noise monitoring, traffic monitoring, air quality monitoring, smart

parking, smart grid, and smart railway control system to solve those problems and

provide a better living environment for the people in cities. More details are given

below:

Smart Grid

IoT accelerates the transformation of the traditional grid to the smart grid with the

capabilities of real-time monitoring, situational awareness, and intelligence control to

make the electric network more sustainable, reliable, secure, and efficient [25]. The IoT

can help to upgrade the grid in two aspects, which are improving the efficiency of the


P a g e | 10

distribution of the energy and the quality of electricity. With the help of advanced

photovoltaic and battery technologies, the energy flow in the smart grid becomes

bidirectional. Thus, smart power meters need to be connected to measure the energy

flow in the grid to meet the users need and improve the efficiency of the energy

distribution in the electricity network. The real-time monitoring of the energy flow can

also guarantee high-quality energy and uninterrupted energy supply in the smart grid

[26].

Smart Parking

The parking spot can be equipped with smart parking meters by using RFID or Near

Field Communication (NFC) to realize the detections and identifications of the cars [10,

27]. There is a novel method to track the car in a parking spot, which is to obtain the

information of the car and the vehicle license plate by using digital camera imaging

sensors. The gathered information of the parking slot can be transmitted to central cloud,

and the user can be easier to find the available parking slots through the smart parking

system which could save the parking time, decrease the CO2 emission, and reduce the

traffic congestion [28].

Environment Monitoring

Environmental issues such as climate changes have received much attention recently,

making it an active and hot research area. The acquisition of the environmental data can

be achieved by a variety of technologies such as remote sensing, geographical

information system, and global positioning system [29]. Wireless Sensor Network can

be another promising technology to collect environmental data automatically, it can be

used together with the cloud computing to collect and analyze the received

environmental data to understand and react to the environment [30]. Internet of things

can provide support for the transmission, storage, and management of a large amount

of data to accurately record the trend of the climate change over the years [31].


P a g e | 11

1.3 Motivation of this work:

1.3.1 Motivations

In recent years, people increasingly draw attention to climate change and the

greenhouse effect. To statistically study the climate change and the impacts of the

human activities on the environment, we need to collect some basic environmental

parameters like rainfall, temperature, CO2, wind speed, and air quality [31]. Sensor

technologies and wireless network integrated with the IoT technology play essential

roles to gather information from the environment. An environmental sensing platform

is an essential tool to collect, transmit, and process environmental parameters to analyze

and monitor climate change. The environmental sensing platform can also be used in

other activities such as structural monitoring, smart agriculture, and asset monitoring.

A structural health monitoring system can be used to make sure the physical

configurations such as buildings and bridges are structurally safe [32]. Some sensors

such as accelerometers and piezoelectric transducers (PZT) can be densely mounted on

the structure to provide health level and spot the damage sites of the architecture [33].

Some signal processing method should be implemented and integrated with the IoT

system to remove the noise generated by the sensors to determine the actual condition

of the structure [34]. Smart farming is the implementation of different sensors and

communication methods to form a wireless sensor network to wirelessly collect

environmental data like temperature, soil humidity, and rainfall [35]. The collected data

can be sent to a cloud for further processing to realize precision farming. Utilizing the

smart farming wireless sensor network, the farmers can improve the production of the

crop and reduce the maintenance fee by adopting corresponding measures based on the

current condition of the land [36]. There are also some valuable assets such as

substations and wind turbines which are directly placed in the outdoor environment.

The surrounding conditions of these outdoor devices should be continuously monitored


P a g e | 12

by sensors to guarantee they are in a safe environment and are prevented from

malfunctions. IoT is the key enabling technology to achieve the monitoring of the

environment and provide useful data for further analyzation and decision making.

1.3.2 Challenges

However, some technical bottlenecks need to be overcome to fully meet the

requirements of the wide and dense deployments of the environmental sensing systems.

The coverage and the energy consumption of the sensor nodes are two main issues that

need to be solved. For some smart city applications such as traffic monitoring and

weather forecasting, the user would like to achieve city-scale monitoring, and the

design of a communication system to cover a whole city can be challenging. It will be

very inconvenient to use wire to form a communication network, and the deployment

cost can also be high. Thus, the wireless communication method will be chosen to build

a network. However, the coverage of the wireless sensor node is limited, and there are

two solutions to overcome this problem. The first is to use the multi-hop

communication technology where each sensor node can act as a vendor to both receive

and transmit the signal to other nodes, and the information can be transmitted to the

gateway in the end. Another option is to use the Low Power Wide Area Network

(LPWAN) technology such as LoRaWAN, the coverage of each gateway can be up to

15-30 km, and only a few gateways can provide coverage over a large region.

Since each node is a wireless sensor node, the battery is usually the primary energy

source to power the node. The energy consumption of the sensor node should be kept

low to prolong the lifespan of the node and guarantee the overall performance of the

whole sensor network. Also, the sensor node can be integrated with energy harvesting

technology to obtain energy from an ambient environment like solar energy, radio

frequency energy, and mechanical energy to become a self-powered device. There are


P a g e | 13

other technologies need to be used in conjunction with the IoT technology to achieve

the full functional environmental sensing platform. A central cloud should be used to

realize the presentation of the measured results of the sensors in real time and the

storage of the measured data. Wireless sensor network technologies should be used to

form a large communication network to collect data from different node deployed over

a wide region.

1.3.3 Objectives

This thesis is aimed to solve the aforementioned research problems with the following

main objectives:

1. To develop an accelerometer-based anemometer with low power consumption

which could be part of the weather station. The anemometer should have a compact

size and can be powered by using the battery to increase the deployment flexibility.

2. To investigate and compare different communication protocols, and to study the

advantages of selected LPWAN protocols in IoT applications. Then, we will build

the wireless sensor network for the environmental sensing platform by

implementing suitable communication technologies.

3. To develop compact, low cost, and robust environmental wireless sensing nodes

with low development and maintenance fee. The sensing node can be easily

deployed and integrated into an environmental wireless sensor network which can

communicate with other nodes or gateways to collect and transmit useful

information.

4. To utilize cloud and IoT based wireless sensor network to develop an environmental


P a g e | 14

sensing platform which could realize real-time monitoring, data presentation, and

data storage.

1.3.4 Original Contributions:

⚫ The first contribution is the development of an ultra-low energy consumption

anemometer. A novel method to measure the wind speed and direction is developed

which uses a 6-axis accelerometer to measure the change of the acceleration

induced by the wind force and then to calculate the corresponding wind speed. The

power consumption of the proposed anemometer is only 3.42 mW, which is

significantly smaller than other anemometers, thus batteries can be used as the

energy source to power the proposed anemometer instead of the power line. The

new design can make an anemometer to be a standalone device powered by a

battery only, which increases the deployment flexibility of the sensor node. The

proposed anemometer has low energy consumption and compact size which also

allows it to be easily integrated to be part of an environmental sensing platform.

⚫ The second contribution of this work is the development of an IoT based

environmental sensing platform which consists of wireless sensor nodes, a LPWA

communication network, data visualization and storage cloud. The environmental

sensing platform has the ability of collect, send, present, and store various

environmental parameters and it has the advantages of easy deployment, real-time

accessibility, low deployment and maintenance cost. The sensing device in the

environmental platform is a true standalone device which has wireless connection

and can be powered by using a solar panel. The data gathered by the sensor node

can be transmitted to a data visualization platform named Cayenne in real time, and

the user can easily access the collected data from sensors deployed in different

areas.


P a g e | 15

1.4 Thesis Outline:

This thesis consists of five chapters, and it aims to provide an environmental sensing

platform composed of different wireless sensor nodes by adopting the concept of IoT.

We aim to reduce the energy consumption of the sensor node to prolong the battery

lifetime of the end device and avoid replacing the battery too often to increase the

flexibility of the deployment of the sensor nodes. The structure of the thesis is organized

as follows.

⚫ Chapter 1 introduces the IoT and discusses the rapid development of IoT

technologies over recent years. It also introduces wide applications of IoT in

different areas and explore how IoT can help us to improve the living quality and

working efficiency. It stresses the importance of the environmental sensing

applications and the role that IoT can play in related fields. The main aims and the

structure of the thesis are also included in this chapter.

⚫ Chapter 2 introduces different types of communication protocols and how wireless

sensors can be connected by using these protocols to form a wireless sensor

network. Various communication protocols can be divided into two parts, which

are the short-range communication protocols and the LPWAN communication

protocols. LPWAN communication methods have merits of wide coverage and

consuming less energy, which is more suitable for IoT applications. WSN is a

sensor network composed of different sensors deployed in various places to gather

useful information for users, and it can be used to construct the environmental

sensing platform.

⚫ Chapter 3 proposes a compact anemometer with low energy consumption which

could be integrated to an environmental sensing platform. The anemometer can

measure both the wind speed and the wind direction by calculating the


P a g e | 16

accelerations obtained from a 6-axis accelerometer. The proposed anemometer has

a compact size and is designed to consume extremely low energy so that it can be

easily integrated into some power constrained weather stations.

⚫ Chapter 4 presents an environmental sensing platform consisted of self-powered

wireless sensor nodes, the wireless sensor network, and data storage cloud.

Different sensors can be easily integrated into the existing sensing platform. The

sensor node is designed to have a sleep mode, and the device can be put into the

sleep mode most of the time to save energy. A 3 W solar panel is used to charge the

battery to make the whole device to be self-powered.

⚫ Chapter 5 summaries major contributions of this thesis and provides future work

recommendations for this research topic.

References:

[1] K. Ashton, "That ‘internet of things’ thing," RFID journal, vol. 22, no. 7, pp.

97-114, 2009.

[2] A. Al-Fuqaha, M. Guizani, M. Mohammadi, M. Aledhari, and M. Ayyash,

"Internet of things: A survey on enabling technologies, protocols, and

applications," IEEE communications surveys & tutorials, vol. 17, no. 4, pp.

2347-2376, 2015.

[3] Lee, C., Park, S., Jung, Y., Lee, Y., & Mathews, M, "Internet of Things:

Technology to Enable the Elderly." In 2018 Second IEEE International

Conference on Robotic Computing (IRC), pp. 358-362, 2018.


P a g e | 17

[4] M. C. J. Thomas, M. Nassiri, D. K. Ismail, and D. K. Ismail, "Internet of things,"

Personal and Ubiquitous Computing, no. 2, pp. 59-61, 2016.

[5] J. Gantz and D. Reinsel, "The digital universe in 2020: Big data, bigger digital

shadows, and biggest growth in the far east," IDC iView: IDC Analyze the future,

vol. 2007, no. 2012, pp. 1-16, 2012.

[6] S. Taylor, "The next generation of the internet," CISCO Point of view, 2013.

[7] M. Z. Shafiq, L. Ji, A. X. Liu, J. Pang, and J. Wang, "A first look at cellular

machine-to-machine traffic: large scale measurement and characterization,"

ACM SIGMETRICS performance evaluation review, vol. 40, no. 1, pp. 65-76,

2012.

[8] S. Ezell, "IoT and smart manufacturing," Inform. Technology and Innovation

Found, 2016.

[9] "2017 Roundup Of Internet Of Things Forecasts.", 2017. [online]. Available:

https://www.forbes.com/sites/louiscolumbus/2017/12/10/2017-roundup-of-

internet-of-things-forecasts/#5f3918991480.

[10] A. Zanella, N. Bui, A. Castellani, L. Vangelista, and M. Zorzi, "Internet of things

for smart cities," IEEE Internet of Things journal, vol. 1, no. 1, pp. 22-32, 2014.

[11] O. Novo, "Blockchain meets IoT: An architecture for scalable access

management in IoT," IEEE Internet of Things Journal, vol. 5, no. 2, pp. 1184-

1195, 2018.

[12] J. Yuan and X. Li, "A Reliable and Lightweight Trust Computing Mechanism

for IoT Edge Devices Based on Multi-Source Feedback Information Fusion,"

IEEE Access, vol. 6, pp. 23626-23638, 2018.

https://www.forbes.com/sites/louiscolumbus/2017/12/10/2017-roundup-of-internet-of-things-forecasts/#5f3918991480

https://www.forbes.com/sites/louiscolumbus/2017/12/10/2017-roundup-of-internet-of-things-forecasts/#5f3918991480


P a g e | 18

[13] C. DeCusatis, Handbook of fiber optic data communication: a practical guide

to optical networking. Academic Press, 2013.

[14] Z. Yang, Y. Yue, Y. Yang, Y. Peng, X. Wang, and W. Liu, "Study and application

on the architecture and key technologies for IOT." In 2011 International

Conference on Multimedia Technology, pp. 747-751, 2011.

[15] L. D. Xu, W. He, and S. Li, "Internet of things in industries: A survey," IEEE

Transactions on industrial informatics, vol. 10, no. 4, pp. 2233-2243, 2014.

[16] L. Antão, R. Pinto, J. Reis, and G. Gonçalves, "Requirements for Testing and

Validating the Industrial Internet of Things." In 2018 IEEE International

Conference on Software Testing, Verification and Validation Workshops

(ICSTW), pp. 110-115, 2018.

[17] "Top Internet of Things Application Areas", 2017. [Online]. Available:

https://www.xtendiot.com/top-internet-things-application-areas/

[18] A. Sun, W. Gong, R. Shea, and J. Liu, "A Castle of Glass: Leaky IoT Appliances

in Modern Smart Homes," IEEE Wireless Communications, vol. 25, no. 6, pp.

32-37, 2018.

[19] R. Mulero et al., "An IoT-aware approach for elderly-friendly cities," IEEE

Access, vol. 6, pp. 7941-7957, 2018.

[20] J. K. Lee, S. N. Robinovitch, and E. J. Park, "Inertial sensing-based pre-impact

detection of falls involving near-fall scenarios," IEEE transactions on neural

systems and rehabilitation engineering, vol. 23, no. 2, pp. 258-266, 2015.

[21] D. Pal, B. Papasratorn, W. Chutimaskul, and S. Funilkul, "Embracing the Smart-

Home Revolution in Asia by the Elderly: An End-User Negative Perception

https://www.xtendiot.com/top-internet-things-application-areas/


P a g e | 19

Modelling," IEEE Access, 2019.

[22] S. M. R. Islam, D. Kwak, M. D. H. Kabir, M. Hossain, and K.-S. Kwak, "The

internet of things for health care: a comprehensive survey," IEEE Access, vol.

3, pp. 678-708, 2015.

[23] G. Yang et al., "A health-IoT platform based on the integration of intelligent

packaging, unobtrusive bio-sensor, and intelligent medicine box," IEEE

transactions on industrial informatics, vol. 10, no. 4, pp. 2180-2191, 2014.

[24] C.-C. Lin and J.-W. Yang, "Cost-efficient deployment of fog computing systems

at logistics centers in industry 4.0," IEEE Transactions on Industrial

Informatics, vol. 14, no. 10, pp. 4603-4611, 2018.

[25] G. Bedi, G. K. Venayagamoorthy, R. Singh, R. R. Brooks, and K.-C. Wang,

"Review of Internet of Things (IoT) in electric power and energy systems,"

IEEE Internet of Things Journal, vol. 5, no. 2, pp. 847-870, 2018.

[26] R. Morello, C. De Capua, G. Fulco, and S. C. Mukhopadhyay, "A smart power

meter to monitor energy flow in smart grids: The role of advanced sensing and

IoT in the electric grid of the future," IEEE Sensors Journal, vol. 17, no. 23, pp.

7828-7837, 2017.

[27] X. Sun and N. Ansari, "Dynamic resource caching in the IoT application layer

for smart cities," IEEE Internet of Things Journal, vol. 5, no. 2, pp. 606-613,

2018.

[28] P. Sadhukhan. "An IoT-based E-parking system for smart cities." 2017

International Conference on Advances in Computing, Communications and

Informatics (ICACCI), pp. 1062-1066, 2017.


P a g e | 20

[29] S. Fang et al., "An integrated system for regional environmental monitoring and

management based on internet of things," IEEE Transactions on Industrial

Informatics, vol. 10, no. 2, pp. 1596-1605, 2014.

[30] M. T. Lazarescu, "Design of a WSN platform for long-term environmental

monitoring for IoT applications," IEEE Journal on emerging and selected topics

in circuits and systems, vol. 3, no. 1, pp. 45-54, 2013.

[31] G. Mois, S. Folea, and T. Sanislav, "Analysis of three IoT-based wireless sensors

for environmental monitoring," IEEE Transactions on Instrumentation and

Measurement, vol. 66, no. 8, pp. 2056-2064, 2017.

[32] J.-L. Lee, Y.-Y. Tyan, M.-H. Wen, and Y.-W. Wu, "Development of an IoT-based

bridge safety monitoring system," 2017 International Conference on Applied

System Innovation (ICASI), pp. 84-86, 2017.

[33] A. Myers, M. A. Mahmud, A. Abdelgawad, and K. Yelamarthi, "Toward

integrating structural health monitoring with internet of things (IoT)," 2016

IEEE International Conference on Electro Information Technology (EIT), pp.

0438-0441, 2016.

[34] M. A. Mahmud, A. Abdelgawad, K. Yelamarthi, and Y. A. Ismail, "Signal

processing techniques for IoT-based structural health monitoring," 2017 29th

International Conference on Microelectronics (ICM), pp. 1-5, 2017.

[35] K. A. Patil and N. R. Kale, "A model for smart agriculture using IoT," 2016

International Conference on Global Trends in Signal Processing, Information

Computing and Communication (ICGTSPICC), pp. 543-545, 2016.

[36] N. Gondchawar and R. S. Kawitkar, "IoT based smart agriculture,"

International Journal of advanced research in Computer and Communication


P a g e | 21

Engineering, vol. 5, no. 6, pp. 2278-1021, 2016.

Chapter 2: Literature Review

P a g e | 22

Chapter 2 Literature Review

2.1 Review of Communication Protocols for IoT Applications

2.1.1 Short-Range Communication Protocols

With the rapid development and wide applications of communication technologies in

the last two decades, it is estimated that more than 26 billion devices will be wirelessly

connected in 2020, and machine-to-machine (M2M) devices will account for nearly

half of the total number of the connected device [1]. Some short-range communication

protocols like Wi-Fi, Bluetooth Low Energy (BLE), and Zigbee play important roles to

realize the IoT vision. This section will give an introduction of the three most widely

used short-range communication protocols and analyze advantages and disadvantages

of these communication methods for IoT applications.

Wi-Fi

Wi-Fi is one of the radio technologies based on IEEE 802.11 family of standards

commonly used for Wireless Local Area Networking (WLAN) of devices promoted by

Wi-Fi Alliance. It operates at 2.4 and 5 GHz and is widely used in both business and

home environment [2]. Wi-Fi can offer supplementary connections for users to the

Internet when the cellular network cannot provide coverage. A typical Wi-Fi router can

provide a maximum 55 m coverage range, and the signal may be degraded by walls and

other complicated environmental factors [3]. Wi-Fi Alliance proposed a new IEEE

802.11ah wireless networking protocol called Wi-Fi HaLow to provide longer

transmission range and obtain better wall penetration performance [4].


P a g e | 23

Bluetooth Low Energy (BLE)

BLE is also known as Bluetooth smart, which is designed and enhanced for short-range

communication with low bandwidth and low latency for IoT applications [5]. It operates

in the ISM bands ranging from 2.400 GHz to 2.4835 GHz and adopts 40 channels with

each having 2 MHz bandwidth [4]. Compared with classic Bluetooth, it has lower

power consumption, shorter setup time, and supporting star network topology with

unlimited numbers of nodes [6, 7]. The energy consumption of a BLE device used for

communication is 28.5 mW.

ZigBee

ZigBee is another short-range wireless network technology created by ZigBee Alliance

based on low-power wireless IEEE 802.15.4 networks standard. The protocol consists

of the physical layer, the Medium Access Control (MAC) layer, the network layer, and

the application layer [8]. There are three types of devices in the ZigBee network layer,

which are the router, the coordinator, and the end device. In a ZigBee network, the end

devices are used to generate the message, the routers have the routing capability, and

the coordinator manages the whole network.

Table 2- 1 Comparison of three short-range communication protocols

Characteristic Wi-Fi ZigBee BLE

Launching year 1997 2003 2010

PHY/MAC IEEE 802.11.1 IEEE 802.15.4 IEEE 805.15.1

Frequency band 2.4 GHz 2.4 GHz 2.4 GHz

Coverage 100 m 10 - 100 m 30 m

Data rate 54 Mbit/s 250 Kbit/s 1 Mbit/s

Topology Star Mesh Star/Mesh

Power level 90-350 mW 72-84 mW 26.5-28.5 mW

Alliance Wi-Fi Alliance ZigBee Alliance Bluetooth SIG


P a g e | 24

2.1.2 Low Power Wide Area Network (LPWAN) Protocols

LPWAN technologies combine both robust modulation and low data rate to achieve

long coverage and low energy consumption. A Low Power Wide Area (LPWA) base

station can cover up to 15 km in rural areas, and the maximum battery life of the LPWA

device can be up to 10 years [9, 10]. There are several LPWAN technologies, such as

SigFox, NB-IoT, and LoRaWAN. SigFox is a single operator network which plans to

offer coverage all over the world incorporation with member companies [11]. However,

it can only obtain a data rate up to 100 b/s in the uplink, and the maximum packet

payload is limited to 12 bytes, which also restricts the wide application of SigFox.

Narrow Band IoT (NB-IoT) is a cellular-based licensed technology introduced by the

3rd Generation Partnership Project (3GPP). NB-IoT can coexist with existed GSM and

LTE networks, and it is operated by telecommunication companies to provide reliable

wireless access for low-power devices. LoRaWAN is different from SigFox and NB-

IoT. LoRaWAN operates in the unlicensed band, which makes it can be used free of

charge. It also allows to set up the private network which can be integrated into different

global network platforms (e.g., The Things Network) [11] which could increase the

flexibility of deployments. The merits of the LoRaWAN put it in an advantage position

compared with other LPWAN technologies.

Table 2- 2 Comparison of three LPWAN communication protocols

Characteristics LoRaWAN SigFox NB-IoT

Frequency (MHz) 433/868/780/915 868 832-862 (Upload)

791-821 (download)

Bandwidth (kHz) 125/250 0.1 180

Data rate upload 0.25 kbps – 50

kbps

100 bps 200 kbps


P a g e | 25

Data rate download 0.25 kbps – 50

kbps

600 bps 200 kbps

Network operator Private operators Sigfox Telecommunication

Companies

Transmission power

consumption

88.4 - 493.2 mW

[12-15]

181.3 – 1036.0

mW [16-18]

479.89 – 1032.3

mW [19]

Receiving power

consumption

40.7 – 170.2

mW [13-15, 20,

21]

31.1 – 199.8 mW

[16-18]

75.1 – 240.1 mW

[19]

Battery life 5-10 years 5-10 years 5-10 years

Developer LoRa Alliance Sigfox 3GPP

Developed Year 2015 2009 2016

Coverage 2-5 km (urban),

15 km (rural)

3-10 km (urban),

17 km (rural)

1 km (urban), 10 km

(rural) [22]

Modulation Chirp Spread

Spectrum

DBPSK/GFSK GMSK/SC-FDMA

Sensitivity (dBm) -142 -142 -142.8 [23]

Topology star Star star

Payload Upload

Length (bytes)

51 [24] 12 128

Payload Download

Length (bytes)

51 8 85

LoRa Physical Layer:

LoRaWAN is a LPWAN protocol for wide area networks based on the long-range LoRa

radios patented by Semtech Corporation [11, 25]. It is designed to achieve long range

and low energy communication between the end devices and gateways. The LoRa

physical layer utilizes Chirp Spread Spectrum (CSS) modulation, which can improve


P a g e | 26

the resilience and robustness against interference, Doppler effect, and multipath [9]. It

operates at 868 MHz in Europe and 915 MHz in the USA with a typical bandwidth of

125 and 250 kHz depending on different spreading factors (SF). An Adaptive Data Rate

(ADR) mechanism is adopted by LoRa to meet the needs of varying transmission

scenarios, and some parameters such as bandwidth, spreading factor and Code Rate

(CR) can be adapted to realize the customization of the LoRa modulation. Spreading

factor is used to adjust the bandwidth and data rate to achieve better coverage. The

spreading factors of the LoRa can also be changed from 7 to 12. Spreading Factor 7 has

the shortest coverage with the highest data rate, and the widest coverage can be

achieved by using spreading factor 12 with the lowest data rate. A detailed comparison

between different spreading factors can be found in Table 2-3.

Table 2- 3 The comparison between different LoRaWAN spreading factors based

on EU 868MHz band

Spreading factor Bandwidth(kHz) Air bit rate

(kbps)

Sensitivity (dBm)

7 250 10.936 -122

7 125 5.468 -124

8 125 3.125 -126

9 125 1.757 -129

10 125 0.976 -132

11 125 0.537 -134.5

12 125 0.293 -137

The data rate of the LoRaWAN transmission can be calculated based on Equation (2.1).

Rb = SF ∗B

2SF ∗ CR (2.1)

where Rb is the data rate, SF is the spreading factor, B is the bandwidth, and CR is the


P a g e | 27

code rate.

LoRaWAN MAC Layer and Network Architecture:

LoRaWAN MAC layer provides the medium access control mechanism on top of the

LoRa physical layer to enable communication between multiple devices and gateways

[11]. It has a star topology where each end nodes are connected to the gateway directly.

LoRaWAN defines three types of devices with different transmission mechanisms

namely Class A, Class B, and Class C. Class A is mandatory for all LoRaWAN devices,

it allows the bidirectional communication between the gateway and the end devices.

Two downlink receiving windows will only be scheduled after a successful uplink

signal interpreted by the gateway. The two downlink receiving windows start 1 and 2s

after the end of the uplink transmission. Class A devices have the lowest energy

consumption among all LoRaWAN devices. Class B devices are synchronized by using

periodic beacon sent by the gateway to arrange an additional receiving window for

downlink transmission. PNG stands for the ping-slot and BCN means the beacon. The

gateway can then be able to send a downlink signal to the end device without prior

successful transmissions. Class C devices will always listen to the gateway unless they

are in the uplink transmission periods. Since Class C devices are always on, they are

the most power-hungry devices compared with other LoRaWAN devices.

Fig. 2- 1 Three classes of LoRaWAN end devices [26].


P a g e | 28

There are two ways to activate end devices in a LoRaWAN network, which are the

Activation By Personalization (ABP) and the Over The Air Activation (OTAA). Three

sets of security keys, namely Network Session Key (NwkSKey), Application Session

Key (AppSKey), and Application Key (AppKey) are used for activating devices during

transmission. All three keys have a length of 128 bits. Network Session Key and

Application Session Key are fixed for a device activated by ABP. OTAA utilizes

Application Key to dynamically derive Network Session Key and Application Session

Key to join the network on every activation. Thus, OTAA is a more secure activation

method compared with ABP. The LoRaWAN architecture (Fig. 2-2) is that the end

device transmits data to a gateway by using LoRaWAN protocol and the gateway is

responsible for furthering forwarding the received message to the network server by

using an IP-based protocol (e.g., Ethernet, Wi-Fi or cellular).

Fig. 2- 2 The LoRaWAN architecture [27].


P a g e | 29

2.2 Review of Wireless Sensor Network

The drastic improvements in micro-electro-mechanical-system (MEMS) technology,

advanced wireless communication methods, and innovative networking protocols

enable forming a network of small, low-price, and powerful wireless sensor nodes. The

Wireless Sensor Network (WSN) can be defined as a network consists of a large number

of sensor nodes composed of the sensing unit, the data processing unit, and the

communication unit. The protocol stack used by the sinks and all sensor nodes is

composed of three planes and five layers, as shown in Fig. 2- 3. The WSN is an essential

IoT technology which can be applied in various areas like military applications,

environmental applications, health applications, smart home applications, and other

commercial applications [28].

Fig. 2- 3 Five layers protocol stacks for wireless sensor network [29].


P a g e | 30

Some sensor nodes may be placed at places hard to reach, and therefore, it will be

difficult to change the battery of the sensor. The energy consumption and power

efficiency are the most important design criteria of a sensor node. Since the power

consumption of the sensor itself is very low, most of the energy is consumed during the

transmission of the signal. It is essential to design a power efficiency transmitting

protocol, which can also maintain the Quality of Service (QoS) and security at the same

time [30]. The deployment of sensor nodes can be formulated as a constrained multi-

objective optimization problem where the aim is to maximize the coverage and the

lifetime and minimize the power consumption and the number of the deployed nodes

[31]. Liu Jingxian proposed that the sensor node can harvest the RF energy from the

sink node by investing the optimal energy beamforming [32]. Samith Abeywickrama

improved this idea by using a novel scheme that facilitates energy beamforming by

utilizing the Received Signal Strength Indicator (RSSI) value to estimate the channel

[33]. Solar energy harvesting technology is a very mature energy harvesting method to

obtain energy in the outdoor environment. The solar energy can be a very promising

method to provide electricity to the sensor nodes due to its relatively high power density

[34]. Instead of gathering energy from other energy sources, some energy saving

techniques can be applied in the communication and networking of the wireless sensor

network. Networking Coding and Power Control based Routing is designed by X, Liu

which can reduce the number of packets transmitted by encoding multiple packages

which have the same destination into one packet and then less energy will be used in

the communication [35].

To provide a reliable and low energy consumption connection for the wireless sensor

node, which may be placed in lots of different places, a low power and long-range

machine-to-machine communication method should be used. The current low-power

and long-range M2M solutions can be basically divided into three parts, which are

LPWA network, IEEE 802.11ah, and cellular-based network infrastructure [36].

LoRaWAN is one of the most widely used LPWA networks especially in Europe. LTE-


P a g e | 31

M and NB-IOT are two typical cellular-based M2M solutions. They can all cover a

wide area and consume less energy, but the data rate is relatively low in compensation.

For IEEE 802.11ah, although it has a higher data rate, it consumes more power and has

less communication coverage. LoRaWAN has a star connected network topology which

combines low data rate and robust modulation to achieve multi-kilometer

communication range, and it is a suitable communication method used to form a private

wireless sensor network [9].

The WSN has very wide applications, and it can be a very promising technology to

collect ambient environmental data, and further transmit, store and analyze the collected

data. It can be used in congestion with other communication technologies, networking

methods, and data processing techniques to improve the overall performance of the

sensing network. Murad Khan developed a smart home control system which not only

saves energy but also is aware of the interference of different wireless technologies

coexisting in the same ISM band [37]. A coordinator is used to receive the signal from

an isolated WSN and transfer those parameters to the management station via power

line communication (PLC) and the control signal for home appliances are also sent

through PLC as well to reduce the impact of wireless interferences [21]. An intelligent

controller by integrating internet of things with cloud computing and web services is

designed to control the heating ventilation and air conditioning (HVAC), the random

neural network is also implanted on the sensor node and base station in this case [38].

The occupancy status can be predicted by using the CO2 sensor measuring the

concentration of CO2 in the air and using some algorithms like machine learning to

predict the existence of people [39, 40].

There are some design criteria to be met to unleash the full functionality of the sensor

nodes in a wireless sensor network. The node should have a compact size, and the

battery will be the primary energy source so that it can be easily deployed without wired

connections [41]. The energy consumption of the end device should be kept low to


P a g e | 32

increase the lifetime of both the node and the wireless sensor network. An energy

harvesting system can be added to the sensor node, and the end device can continuously

work without changing batteries to prolong the lifetime of the wireless sensor network

[42]. There are some restrictions on the size and the cost of the sensor node so that it

can be densely deployed. Also, each device should be easily connected to the cloud,

and the users can have access to the data in real time. The interoperability of the wireless

sensor network is another critical design criterion so that more functions and other

devices can be easily integrated into the existed sensing platforms.

2.3 Review of Environmental Monitoring Activities

Some essential environmental parameters such as temperature, humidity, brightness,

wind speeds, and wind directions are needed in different applications such as

environmental monitoring, asset monitoring, smart agriculture, weather forecasting,

aviation, and load prediction. 40 percent of the aviation accidents are due to the adverse

weather conditions, and therefore, some weather stations should be deployed near the

airport to know the real-time weather conditions [43]. To reduce the greenhouse effect

by using renewable energy to decrease the emission of CO2, the wind turbine can be a

sustainable replacement of the thermal power plant. The site selection of the wind farm

is very crucial to guarantee more energy can be obtained from the environment, and

wind sensors can be deployed in different sites to record the wind speeds and directions

over a long time. The wind turbines can then be installed in suitable places with

optimized facing directions which could maximize the energy extracted from the wind

[44]. Weather prediction is essential for our day by day life, especially in agribusiness,

and national weather data may not contain the precise information of some particular

areas [45]. It is essential to deploy weather stations to monitor and predict the weather

conditions near the farm to realize precision agriculture.


P a g e | 33

The environmental monitoring system also plays an important role in the electricity

distribution system to improve the safety and stability of the power systems. The main

contribution of the environmental sensing platform to the power grid can be categorized

into two parts which are the condition monitoring of the outdoor devices in the grid and

the load predictions based on the weather conditions. Since some important assets in

the power grid such as transformers, substations, and power lines are directly placed in

the outdoor environment. The conditions of these devices should be continuously

monitored so that it will be easy to do the maintenance work and prevent them from

malfunctions to cause interruptions in the power supply [46-48]. Weather conditions

have an important influence on the generation, transmission, and the distribution of

electric power [48, 49]. The clean and environmentally friendly renewable energy such

as photovoltaic and wind power generation are now penetrating the power grid at an

accelerating speed [48, 50]. However, the randomness and volatility of renewable

energy also affect the safety and stability of the power system, and weather stations

should provide abundant, accurate and real-time metrological data to predict the load

and achieve the smart control in the modern power grid [46].

Currently, with the help of advanced communication methods and sensing technologies,

a weather station can be integrated with more functions which can collect, process, and

transmit various environmental parameters. By adopting the concept of the wireless

sensor network, the traditional weather station can be transformed from a standalone

device to be part of an environmental sensing platform. The weather stations can be

deployed in different places, including in harsh environments such as high mounts,

deserts, and Antarctica [51]. The data can be automatically uploaded to the central cloud

of the environmental sensing platform without the presence of the human. By using the

cloud computing technology, the environmental sensing platform can then present and

process the data, and the user can have real-time access to the measured data and make

decisions based on the collected information [52]. A large number of research projects

such as Smart Santander and URBAN GreenUP aims to continuously monitor


P a g e | 34

environment by using advanced IoT technologies to provide environmental, social and

economic value.

Fig. 2- 4 presents a typical weather station, including the wind speed sensor, the wind

direction sensor, the rain bucket, solar panels, and wires to connect different devices

[47]. The weather station has a bulky size which may make it to be easily spotted and

difficult to be deployed. It can also be found that some connection wires are directly

exposed to the environment, and the wires may be damaged by the birds or strong winds,

which could increase the chance of malfunctions.

Fig.2- 4 A weather station mounted on the adjustable tripod [44].

There are some design criteria need to be met to unleash the full functionality of the

WSN based environmental platform. The sensor node should be small in size and is

powered by the battery so that it can be easily deployed without wired connections [53].


P a g e | 35

The energy consumption of the nodes should be kept low to increase the lifetime of

both the node and the wireless sensor network. An energy harvesting system can be

added to the sensor node, and the end device can continuously work without changing

batteries to prolong the lifetime of the whole environmental monitoring system [54].

There are some restrictions on the size and the cost of the weather station so that it can

be densely deployed. Also, the weather station should be easily connected to the cloud,

and the users can have access to the data in real time. The interoperability of the weather

station is another critical design criterion so that more functions and other devices can

be easily integrated into the existed sensing platforms.

2.4 Summary

A detailed introduction of the popular communication protocols, including both the

short-range communication protocols and the LPWAN protocols was given. The studies

and applications of the Wireless Sensor Network have been discussed as well. It can be

concluded that LPWAN protocols have the merits of wide coverage and low energy

consumption, which is more suitable to transmit data in an energy-constrained wireless

sensor network. Three popular LPWAN protocols, which are the LoRaWAN, SigFox,

and NB-IoT are studied, and the LoRaWAN has the advantages of flexible private

deployment and low energy consumption feature. The information in this chapter

provides basic knowledge of the communication protocols and the wireless sensor

networks which may be needed in the following sections of the thesis. This chapter also

spots the research gaps and challenges of the current weather stations for us to

overcome in this work.


P a g e | 36

References:

[1] P. Burzacca, M. Mircoli, S. Mitolo, and A. Polzonetti, "“iBeacon” technology

that will make possible Internet of Things," 2014.

[2] P. S. Henry and H. Luo, "WiFi: what's next?," IEEE Communications Magazine,

vol. 40, no. 12, pp. 66-72, 2002.

[3] V. Radu, L. Kriara, and M. K. Marina, "Pazl: A mobile crowdsensing based

indoor WiFi monitoring system," Proceedings of the 9th International

Conference on Network and Service Management (CNSM 2013), pp. 75-83,

2013.

[4] N. Xia, H.-H. Chen, and C.-S. Yang, "Radio resource management in machine-

to-machine communications—A survey," IEEE Communications Surveys &

Tutorials, vol. 20, no. 1, pp. 791-828, 2018.

[5] S. Al-Sarawi, M. Anbar, K. Alieyan, and M. Alzubaidi, "Internet of Things (IoT)

communication protocols," 2017 8th International Conference on Information

Technology (ICIT), pp. 685-690, 2017.

[6] F. Samie, L. Bauer, and J. Henkel, "IoT technologies for embedded computing:

A survey," Proceedings of the Eleventh IEEE/ACM/IFIP International

Conference on Hardware/Software Codesign and System Synthesis, p. 8, 2016.

[7] T. Salman, "Internet of things protocols and standards," ed: Academic Press,

2015.

[8] L. Atzori, A. Iera, and G. Morabito, "The internet of things: A survey,"

Computer networks, vol. 54, no. 15, pp. 2787-2805, 2010.

[9] F. Adelantado, X. Vilajosana, P. Tuset-Peiro, B. Martinez, J. Melia-Segui, and


P a g e | 37

T. Watteyne, "Understanding the limits of LoRaWAN," IEEE Communications

magazine, vol. 55, no. 9, pp. 34-40, 2017.

[10] Y. Song, J. Lin, M. Tang, and S. Dong, "An Internet of energy things based on

wireless LPWAN," Engineering, vol. 3, no. 4, pp. 460-466, 2017.

[11] J. Haxhibeqiri, E. De Poorter, I. Moerman, and J. Hoebeke, "A survey of

lorawan for iot: From technology to application," Sensors, vol. 18, no. 11, p.

3995, 2018.

[12] J. Petäjäjärvi, K. Mikhaylov, R. Yasmin, M. Hämäläinen, and J. Iinatti,

"Evaluation of LoRa LPWAN technology for indoor remote health and

wellbeing monitoring," International Journal of Wireless Information Networks,

vol. 24, no. 2, pp. 153-165, 2017.

[13] B. Kim and K.-i. Hwang, "Cooperative downlink listening for low-power long-

range wide-area network," Sustainability, vol. 9, no. 4, p. 627, 2017.

[14] G. Conus, G. Lilis, N. A. Zanjani, and M. Kayal, "An event-driven low power

electronics for loads metering and control in smart buildings," 2016 Second

International Conference on Event-based Control, Communication, and Signal

Processing (EBCCSP), pp. 1-7, 2016.

[15] P. Neumann, J. Montavont, and T. Noël, "Indoor deployment of low-power wide

area networks (LPWAN): A LoRaWAN case study," 2016 IEEE 12th

International Conference on Wireless and Mobile Computing, Networking and

Communications (WiMob), pp. 1-8, 2016.

[16] C. Gomez, J. C. Veras, R. Vidal, L. Casals, and J. Paradells, "A Sigfox energy

consumption model," Sensors, vol. 19, no. 3, p. 681, 2019.


P a g e | 38

[17] E. Morin, M. Maman, R. Guizzetti, and A. Duda, "Comparison of the device

lifetime in wireless networks for the internet of things," IEEE Access, vol. 5, pp.

7097-7114, 2017.

[18] D. M. Hernandez, G. Peralta, L. Manero, R. Gomez, J. Bilbao, and C. Zubia,

"Energy and coverage study of LPWAN schemes for Industry 4.0," 2017 IEEE

International Workshop of Electronics, Control, Measurement, Signals and their

Application to Mechatronics (ECMSM), pp. 1-6, 2017.

[19] M. Lauridsen, R. Krigslund, M. Rohr, and G. Madueno, "An Empirical NB-IoT

Power Consumption Model for Battery Lifetime Estimation," 2018 IEEE 87th

Vehicular Technology Conference (VTC Spring), pp. 1-5, 2018..

[20] C. A. Socarrás Bertiz, J. J. Fernández Lozano, J. A. Gomez-Ruiz, and A. García-

Cerezo, "Integration of a Mobile Node into a Hybrid Wireless Sensor Network

for Urban Environments," Sensors, vol. 19, no. 1, p. 215, 2019.

[21] K. Mikhaylov and J. Petäjäjärvi, "Design and implementation of the plug&play

enabled flexible modular wireless sensor and actuator network platform," Asian

Journal of Control, vol. 19, no. 4, pp. 1392-1412, 2017.

[22] K. Mekki, E. Bajic, F. Chaxel, and F. Meyer, "Overview of Cellular LPWAN

Technologies for IoT Deployment: Sigfox, LoRaWAN, and NB-IoT," 2018

IEEE International Conference on Pervasive Computing and Communications

Workshops (PerCom Workshops), pp. 197-202, 2018.

[23] R. Ratasuk, B. Vejlgaard, N. Mangalvedhe, and A. Ghosh, "NB-IoT system for

M2M communication," 2016 IEEE wireless communications and networking

conference, pp. 1-5, 2016.

[24] B. Vejlgaard, M. Lauridsen, H. Nguyen, I. Z. Kovács, P. Mogensen, and M.


P a g e | 39

Sorensen, "Coverage and capacity analysis of sigfox, lora, gprs, and nb-iot,"

2017 IEEE 85th vehicular technology conference (VTC Spring), pp. 1-5, 2017.

[25] M. Rizzi, P. Ferrari, A. Flammini, and E. Sisinni, "Evaluation of the IoT

LoRaWAN solution for distributed measurement applications," IEEE

Transactions on Instrumentation and Measurement, vol. 66, no. 12, pp. 3340-

3349, 2017.

[26] T. Bouguera, J.-F. Diouris, J.-J. Chaillout, R. Jaouadi, and G. Andrieux, "Energy

consumption model for sensor nodes based on LoRa and LoRaWAN," Sensors,

vol. 18, no. 7, p. 2104, 2018.

[27] "Background Information about LoRaWAN." [Online]. Available:

https://www.thethingsnetwork.org/docs/lorawan/

[28] E. H. Callaway Jr, Wireless sensor networks: architectures and protocols. CRC

press, 2003.

[29] I. F. Akyildiz, W. Su, Y. Sankarasubramaniam, and E. Cayirci, "Wireless sensor

networks: a survey," Computer networks, vol. 38, no. 4, pp. 393-422, 2002.

[30] M. Saikia, U. K. Das, and M. A. Hussain, "Secure energy aware multi-path

routing with key management in wireless sensor network," 2017 4th

International Conference on Signal Processing and Integrated Networks (SPIN),

pp. 310-315, 2017.

[31] S. Sengupta, S. Das, M. D. Nasir, and B. K. Panigrahi, "Multi-objective node

deployment in WSNs: In search of an optimal trade-off among coverage,

lifetime, energy consumption, and connectivity," Engineering Applications of

Artificial Intelligence, vol. 26, no. 1, pp. 405-416, 2013.

https://www.thethingsnetwork.org/docs/lorawan/


P a g e | 40

[32] J. Liu, K. Xiong, P. Fan, and Z. Zhong, "RF energy harvesting wireless powered

sensor networks for smart cities," IEEE Access, vol. 5, pp. 9348-9358, 2017.

[33] S. Abeywickrama, T. Samarasinghe, and C. K. Ho, "Wireless energy

beamforming using signal strength feedback," 2016 IEEE Global

Communications Conference (GLOBECOM), pp. 1-6, 2016.

[34] C. Alippi and C. Galperti, "An adaptive system for optimal solar energy

harvesting in wireless sensor network nodes," IEEE Transactions on Circuits

and Systems I: Regular Papers, vol. 55, no. 6, pp. 1742-1750, 2008.

[35] X. Tian, Y.-H. Zhu, K. Chi, J. Liu, and D. Zhang, "Reliable and energy-efficient

data forwarding in industrial wireless sensor networks," IEEE Systems Journal,

vol. 11, no. 3, pp. 1424-1434, 2017.

[36] H. Wang and A. O. Fapojuwo, "A survey of enabling technologies of low power

and long range machine-to-machine communications," IEEE Communications

Surveys & Tutorials, vol. 19, no. 4, pp. 2621-2639, 2017.

[37] M. Khan, B. N. Silva, and K. Han, "Internet of things based energy aware smart

home control system," IEEE Access, vol. 4, pp. 7556-7566, 2016.

[38] A. Javed, H. Larijani, A. Ahmadinia, and D. Gibson, "Smart random neural

network controller for HVAC using cloud computing technology," IEEE

Transactions on Industrial Informatics, vol. 13, no. 1, pp. 351-360, 2017.

[39] S. H. Ryu and H. J. Moon, "Development of an occupancy prediction model

using indoor environmental data based on machine learning techniques,"

Building and Environment, vol. 107, pp. 1-9, 2016.

[40] T. H. Pedersen, K. U. Nielsen, and S. Petersen, "Method for room occupancy


P a g e | 41

detection based on trajectory of indoor climate sensor data," Building and

Environment, vol. 115, pp. 147-156, 2017.



[42] S. Priya and D. J. Inman, Energy harvesting technologies. Springer, 2009.

[43] J. McCarthy, "Advances in weather technology for the aviation system,"

Proceedings of the IEEE, vol. 77, no. 11, pp. 1728-1734, 1989.

[44] D. A. Aponte-Roa, L. B. Montalvan, C. Velazquez, A. A. Espinoza, L. F.

Velazquez, and R. Serrano, "Evaluation of a low-cost, solar-powered weather

station for small-scale wind farm site selection," 2018 IEEE International

Instrumentation and Measurement Technology Conference (I2MTC), pp. 1-5.

IEEE, 2018.

[45] M. I. Haque, A. H. M. D. Shatil, A. N. Tusar, M. Hossain, and M. H. Rahman,

"Renewable Powered Portable Weather Update Station," 2019 International

Conference on Robotics, Electrical and Signal Processing Techniques

(ICREST), pp. 374-377, 2019.

[46] H. Li, M. K. Ochani, H. Zhang, and L. Zhang, "Design of micro-automatic

weather station for modern power grid based on STM32," The Journal of


[47] M. Kusriyanto and A. A. Putra, "Weather Station Design Using IoT Platform

Based On Arduino Mega," 2018 International Symposium on Electronics and

Smart Devices (ISESD), pp. 1-4, 2018.

[48] P. Dehghanian, B. Zhang, T. Dokic, and M. Kezunovic, "Predictive risk


P a g e | 42

analytics for weather-resilient operation of electric power systems," IEEE

Transactions on Sustainable Energy, vol. 10, no. 1, pp. 3-15, 2019.

[49] G. Iannarelli and C. Boccaletti, "Economic impact of investments in weather

forecasts for distribution system operators: the Italian case," IET Generation,

Transmission & Distribution, vol. 12, no. 20, pp. 4551-4556, 2018.

[50] H. Sangrody, M. Sarailoo, N. Zhou, N. Tran, M. Motalleb, and E. Foruzan,

"Weather forecasting error in solar energy forecasting," IET Renewable Power

Generation, vol. 11, no. 10, pp. 1274-1280, 2017.

[51] D. Cesarini, L. Cassano, M. Kuri, V. Bilas, and M. Avvenuti, "AENEAS: An

energy-aware simulator of automatic weather stations," IEEE Sensors Journal,

vol. 14, no. 11, pp. 3932-3943, 2014.

[52] L. Jiang, L. Da Xu, H. Cai, Z. Jiang, F. Bu, and B. Xu, "An IoT-oriented data

storage framework in cloud computing platform," IEEE Transactions on

Industrial Informatics, vol. 10, no. 2, pp. 1443-1451, 2014.




Chapter 3: A Compact Anemometer with Ultra-low Energy Consumption

P a g e | 43

Chapter 3 A Compact Anemometer with Ultra-

low Energy Consumption

Wind speeds and directions are essential environmental parameters, and it is required

in different environmental monitoring applications such as the aviation, the site

selection of the wind farm, structural health monitoring, etc. However, the energy

consumption of the traditional wind sensors such as the ultrasonic wind sensors and the

hot-film wind sensors typically consumes a large amount of power. These wind sensors

are normally powered by using power lines, and it will be very challenging to use

batteries to power them. Since energy consumption is one of the constrained design

factors needs to be concerned in the development of wireless weather stations. In this

chapter, we would like to propose a novel wind sensor which can measure both the

wind speeds and the wind directions by using a 6-axis accelerometer. The proposed

wind sensor has a compact size and extremely low energy consumption so that it can

be easily deployed and powered by using batteries.

3.1 Introduction

Wind measurement has been widely used in different applications, such as weather

forecasting and outdoor asset monitoring [1]. With the development of advanced

communication technologies, an anemometer can act as a sensor node to be connected

to the internet so that all the measured data can be uploaded and visualized in real time.

To realize the concept of IoT, we need to pay more attention to the requirements of the

wind sensors such as miniaturized size, low cost, and low energy consumption [2, 3].

There are various types of traditional anemometers, such as cup-type anemometers,

ultrasonic anemometers, and thermistors-based anemometers as displayed in Fig. 3-1,


P a g e | 44

Fig. 3-2, and Fig. 3-3. However, none of these existing measuring methods can meet all

three requirements. In this chapter, we will propose a novel and compact anemometer

which can measure both the wind speeds and directions by using a 6-axis accelerometer

with low cost and ultra-low energy consumption.

Fig. 3- 1 A cup-type anemometer.


P a g e | 45

Fig. 3- 2 An ultrasonic anemometer [11].

The cup-type anemometer is the most widely used anemometer because it has low

production costs and can sustain a variety of harsh environments [4]. However, the

implementation and the application of the cup-type anemometer are limited by its bulky

size and mechanical performance degradation over a long time [5-7]. The traditional

cup-type anemometer can only measure the wind speed in one direction, and an extra

part named wind vane is needed to obtain the wind direction. The inertia of the rotating

part will cause measurement errors because the cup will keep rotating when the wind

ceases [4]. To overcome these limitations of the cup type anemometer, some advanced

anemometers like the ultrasonic anemometer and the hot-wire anemometer are

developed. For an ultrasonic anemometer, although the size is shrunken to the

dimension of a few tens of centimetres and there is no moving part. Further

miniaturization will be impractical due to the increasing measurement errors [8, 9]. The

physical structures to mount the ultrasonic transmitters and receivers can severely

interfere with the wind flow when the wind comes from the direction which is parallel


P a g e | 46

to the mounting structures [10, 11]. Also, the application of ultrasonic anemometer will

be restricted due to its high cost and high energy consumption [12, 13]. The temperature

of the air can also affect the travel speed of the ultrasonic wave and therefore affect the

wind speed measurement.

Fig. 3- 3 A hot-wire anemometer [16].

Hot-wire anemometers are developed to measure the drift of heat induced by the wind

and calculating the corresponding wind speed [14, 15]. Fig. 3-3 presents a typical hot-

wire anemometer including the sensing unit and the data processing unit. Although the

size of the sensor has been significantly miniaturized, there are still some limitations

such as the temperature of the ambient environment can affect the measuring results

[16]. Since the film should be heated, and a large amount of energy needs to be

dissipated to conduct the measurement [17, 18]. The energy consumption of a hot-wire

anemometer can be 2 W to elevate the temperature of the sensing head to a desirable

level [19]. There are other wind measuring methods such as remote sensing techniques

[20, 21]. Coherent Doppler LiDAR can be used to measure the wind speed remotely

over a large range without contact with the moving air [22, 23]. Lidar can emit laser


P a g e | 47

light and detect the doppler shift in the backscattered light, and the wind speed can be

measured according to the doppler shift. However, this method cannot obtain accurate

wind speed in specific sites.

The energy consumption of conventional anemometer could be 2 W and some well-

designed low energy consumption anemometer could still consume more than 20 mW

power. The battery level of a wireless weather station could be drained quickly if the

high energy consumption anemometer is integrated into the weather station. In this

chapter, we will introduce a novel, compact, low cost, and low energy consumption

anemometer which can also be integrated as a part of a weather station by using

Bluetooth Low Energy (BLE). The energy consumption of the proposed anemometer

has been reduced to 3.42 mW. In this design, we will calculate the wind speed by

measuring the change of the position of an accelerometer induced by the wind force.

The measured data can be easily processed by using a Gaussian-weighted moving

average filtering algorithm to reduce the error and uncertainty of the measurement.

3.2 Model Establishment and Theory

3.2.1 Sensor Design and Measuring Method

In this design, we propose an accelerometer-based anemometer which can measure the

2-dimensional wind speed by calculating corresponding accelerations of the sensor case

along two axes in a horizontal plane. The new anemometer mainly consists of four parts

which are an accelerometer, a 3D printed sensor case, a universal joint, and a mounting

structure. The accelerometer as shown in Fig. 3-4a is placed in the sphere sensor case

as shown in Fig. 3-4b and 3-4c, and the sphere sensor case is placed under a clamp

stand by using a universal joint connected with a 3D printed pole as presented in Fig.

3-4d. When a wind occurs and hits the sensor case, the sensor case moves corresponding

to the wind. The wind can generate a force exerted on the sensor case, and the sensor


P a g e | 48

case will move towards a new equilibrium position. The new equilibrium position will

result in the changes of accelerations measured by the accelerometer, and the wind

speed can be obtained by calculating the changes of accelerations.

Fig. 3-4 The accelerometer, sensor case, and mounting structure. (a) The

accelerometer; (b) The accelerometer placed in the sensor case; (c) The sphere

sensor case and the universal joint; (d) The sensor case hung over the clamp stand.

Accelerometer

A 6-axis Inertial Measurement Unit (IMU) BMI160 which is designed and developed

by Bosch is used to measure the acceleration. BMI160 has a minimum division of the

0.000598 m/s2 which is sensitive enough to determine the small change of position

induced by the wind. The power consumption of the BMI 160 is only 3.42 mW when

it is in full operation and 10.8 ㎼ when it is in the low-power mode. Thus, to use the

IMU BMI160 to measure the wind speed can significantly reduce the energy

a b

c d


P a g e | 49

consumption compared with other conventional anemometers. The BMI160 is

integrated into a small Printed Circuit Board (PCB) with a size of 30mm * 30mm, as

shown in Fig. 3-4a and it will be placed flat into the sensor case as shown in Fig. 3-4b.

The board also has a central processing unit to process the data and Bluetooth module

to transmit the measured data wirelessly. It can measure three sets of accelerations along

three perpendicular axes. In this design, we only focus on the accelerations along two

perpendicular axes in the horizontal plane, namely Ax and Ay.

3D Printed Spherical Sensor Case

The sensor case is designed to be a sphere with a diameter of 5 cm so that it is perfectly

symmetrical in all directions. And the same amount of force can be exerted on the

sensor case when winds with same speeds from different directions hit the sensor case.

The sensor case is 3D printed by using Acrylonitrile Styrene Acrylate (ASA) which is

thermoplastic material. This material is suitable for outdoor weather applications due

to its high UV stability and chemical resistance. The sensor case consists of two parts

which are the base and the lid. There is a square container embedded in the base half

sphere, and it is used to hold the accelerometer. There is a hole with 3mm in diameter

in the lid which is used to connect with the universal joint. The picture of the base and

the lid can be found in Fig. 3-4b. The structure is designed to have no gaps between the

lid and the base so that the sensor case can prevent the rain and the dust.

Universal Joint and Mounting Structure

A stainless universal joint with low friction is chosen to connect the sensor case to the

mounting structure. It allows the sensor case to move freely along two perpendicular

axes in the horizontal plane. One side of the universal joint is connected to the lid of

the sensor case, and the other side of the joint is connected to a 3D printed pole. Then,

the 3D printed pole is clamped by the clamp stand to fix the anemometer. The main

body of the clamp stand is 15cm away from the anemometer to minimize the wind

interferences it may induce. The whole structure of the sensor case clamped by the


P a g e | 50

clamp stand is shown in Fig. 3-4d. The anemometer can also be easily placed under

other places depends on the application and deployment.

3.2.2 Model Establishment

Wind Force Model:

The wind is the motion of the air, and it can generate a drag force on the object, namely

the wind force. Since the area of the contact surface is small, it can be assumed that

there is no turbulence, and the air pressure is evenly distributed across the contact

surface. Also, the average wind force acting on the sensor case within a short time can

be viewed as a stationary force [1, 2]. The wind speed can be calculated by measuring

the wind force exerted on the sensor case. The wind force F(t) can be expressed as a

function of wind speed varies with time which is proportional to the area of the contact

surface A as well as the density of the air ρ , and it is 1.226 kg/m3 at the standard

atmosphere. The geometry shape of the contact surface can also affect the wind force

generated on the object, and the shape factor is defined as μ. The wind force can be

expressed in the following equation.

F(t) = 1

2μρAV(t)2 (3.1)

where V(t) is the speed of the wind varying with time.

3D Dynamic Model of the Anemometer

1. Top View: Force Analysis

A wind force F(t) as a function of t is generated on the sphere sensor case as shown

in Figure. 3-5. θ(t) is defined as the direction of the wind, and the wind direction is

also a function that varies with time. The diagram of the force analysis from the top

view can be found in Figure. 3-5a. The wind force can then be divided into two parts

along the X-axis and Y-axis. The equations are shown below.


P a g e | 51

Fx(t) = F(t) cos(θ(t)) (3.2)

Fy(t) = F(t) sin(θ(t)) (3.3)

a

b


P a g e | 52

Fig. 3- 5 The 3D force analysis of the anemometer. (a) The wind force divided into

X-axis and Y-axis from the top view of the anemometer; (b) The force analysis

along X-axis and Z-axis; (c) The force analysis along Y-axis and Z-axis.

2. Side View: Equation of Rotation Movement

The anemometer has two Degrees of Freedom (DoFs), and it can do pendulum

movements in the X axis and Y axis. The mass distribution of the sensor case and the

accelerometer is measured, and the mass of the whole structure can be equivalently

replaced by a single point, which is called the centre of mass. The model can be further

simplified as the centre of mass rotating around the universal joint. The wind can exert

a force on the whole contact surface of the sensor case, and an equivalent wind force

exerted point is introduced to simplify the wind force exertion. Then, the wind force

can be viewed as being generated on a single point rather than the whole contact surface.

The sensor case will rotate around the joint when a wind comes, φx(t) and φy(t) are

the rotation angle in X-axis and Y-axis, respectively. There are two torques applied to

the system which are the gravity torque and the torque generated by the wind. The

equations of dynamics of the anemometer moving along two axes can be described in

the following equations.

c


P a g e | 53

Id2φx(t)

dt2 = Fx(t)l1 cos φx(t) − mgl2 sin φx(t) (3.4)

Id2φy(t)

dt2 = Fy(t)l1 cos φy(t) − mgl2 sin φy(t) (3.5)

where I is the moment of inertia of the anemometer, l1 is the distance between the

joint and the wind force exerted point, and l2 is the distance between the centre of

mass and the joint.

3. Acceleration Interpretation

𝜑𝑥(𝑡) and 𝜑𝑦(𝑡) are measured by calculating the changes of accelerations along X-

axis and Y-axis. If there is no wind, the accelerometer will remain in the horizontal

plane and the accelerations of the two axes Ax and Ay are 0 m/s2. When a wind

occurred, the anemometer can be rotated to a new equilibrium position, and the changes

of accelerations can be used to determine the rotation angles. When a rotation occurred,

the gravity will have two components of forces along the two axes of the accelerometer.

Accordingly, the measured accelerations Ax and Ay can be used to determine the

rotation angle φx(t) and φy(t). The following equations give the relation between

the Ax, Ay, φx(t) and φy(t).

Ax(t) = g sin(φx(t)) (3.6)

Ay(t) = g sin (φy(t)) (3.7)

Based on equation (3.1) - (3.7) it can be derived that:

V(t)2 cos(θ(t)) = 2

μρ Al1cos(sin−1(Ax(t)

g))

[Id2(sin−1(

Ax(t)

g))

dt2 + ml2Ax(t)] (3.8)

V(t)2 sin(θ(t)) = 2

μρ Al2cos(sin−1(Ay(t)

g))

[Id2(sin−1(

Ay(t)

g))

dt2 + ml2Ay(t)] (3.9)


P a g e | 54

We can represent the wind speed V(t) and the direction θ(t) of the wind by the

measured accelerations along two axes of the accelerometer.

Maximum and Minimum Wind Speed Analysis

The minimum wind speed measured by the anemometer depends on the sensitivity of

the accelerometer. The sensitivity of the accelerometer is 6.10*10-5 g, and according to

the equation 3.1, the corresponding minimum wind speed to be measured is calculated

to be 0.22 m/s. The measurement of the maximum wind speed depends on the maximum

rotating angle away from the initial point. According to the physical structure of the

sensor box and the universal joint. The maximum rotating angle is considered to be

45° , and the corresponding force induced by the wind acted on the sensor box is equal

to the gravity acted on the sensor box. The maximum wind speed could be measured is

28.26 m/s based on equation 3.1.

3.3 Experimental Results and Discussions

Fig. 3- 6 The setup of the wind speed measurement.

To verify the performance of the proposed accelerometer-based anemometer, we place

the anemometer near to a wind source. The wind source can generate steady winds with


P a g e | 55

different speeds varying from 1 m/s to 10 m/s. A fan-type anemometer is placed

between the wind source and the proposed anemometer to measure the ground truth

value of the wind speed generated by the wind source, as shown in Fig. 3-6. A total

number of 50 sets of accelerations can be measured by the proposed anemometer in 1

second. Based on the collected accelerations in two axes, both the wind speeds and the

wind directions can be derived. The experiment is carried out based on the procedure,

as shown in Table 3-1, and the raw data obtained from the accelerometer is shown in

Fig. 3-7. The measured accelerations will be transmitted to a smartphone by using

Bluetooth, and the measured acceleration will be further processed to obtain the wind

speeds and wind directions.

Table 3- 1 The measurement procedure of the proposed anemometer.

Time period (s) Wind speed (m/s)

0-5 0

5-15 5.5

15-25 5.0

25-35 4.5

35-45 5.5

45-50 0

Fig. 3- 7 The measured accelerations of two axes without data processing.


P a g e | 56

Fig. 3- 8 The measured accelerations of two axes with data processing.

Fig. 3- 9 The comparison between the measured wind velocity and the actual wind

velocity.

From the measured accelerations in the two axes, it can be found that the noise level is

very high, and a filtering algorithm should be applied to remove the noise. A Gaussian-

weighted moving average filtering method is applied to remove the noise. The

measuring frequency of the accelerometer is 50 Hz, and the algorithm can sum up every

30 discrete points and calculate the average value of the wind speed. The averaged

values can then be acted as the modified results, which are more closed to the real values

of the accelerations in those conditions. The results of the processed accelerations after


P a g e | 57

the filtering algorithm are shown in Fig. 3-8 It can be found that the noise has been

removed and the result is more accurate. Then, the wind speeds and the directions can

be calculated based on Equations (3.8) and (3.9), which are shown in Fig. 3-9. Both the

measured results of the wind speed and the wind directions matched well with the actual

values.

To further verify the performance of the proposed anemometer, we have compared the

proposed anemometer with a commercial cup-type anemometer in an open parking lot

as displayed in Fig. 3-10. The proposed anemometer is mounted on the pole near to the

cup-type anemometer to ensure both devices are exposed to the same wind speed and

direction. The upper part of the cup-type anemometer is a wind vane which is used to

measure the wind direction and the lower part of the anemometer has three rotating

wind cups to measure the wind speed.

Fig. 3- 10 The comparison between the proposed anemometer and the

commercial cup-type anemometer.


P a g e | 58

The wind speed and direction were sampled every minute in the time frame. The

weather condition was partly cloudy as shown in Fig. 3-10. The measured wind speed

and direction of two anemometers in 15 minutes are displayed in Figs. 3-11 and 3-12.

Based on the measured results, it can be found that both the wind speed and direction

measured by the two devices are consistent with each other, and errors are constrained

in a reasonable region.

Fig. 3- 11 The wind speed measurements of the commercial anemometer and the

proposed anemometer.

Fig. 3- 12 The wind direction measurements of the commercial anemometer and

the proposed anemometer.


P a g e | 59

The accelerometer is held in the sphere 3D printed sensor case, and the diameter of the

sensor case is only 5 cm. Compared with other anemometers, the proposed anemometer

has a very compact size, which allows it to be more easily deployed. Since the

accelerometer is the only device used to measure the wind speeds and wind directions,

and the power consumption of the anemometer will be the energy consumed by the

anemometer. The energy consumption of the accelerometer is only 3.42 mW, which is

significantly smaller than other wind sensors listed in Table 3-2. The following table

summarizes the comparison of key characteristics between different anemometers and

the proposed anemometer, and it can be concluded that the proposed anemometer has a

compact size with lowest power consumption which is more suitable for IoT

environmental sensing applications.

Table 3- 2 Comparison of proposed anemometer with other designs

Ref

(year)

Figure Anemometer

type

Power

consumption

Size

[24]

2018

12-pressure-

sensors-based

anemometer

400 mW 100 mm in

diameter

[16]

2017

Hot-film

anemometer

20 mW Not given

[25]

2018

Hot-film

anemometer

20 – 45 mW Not given


P a g e | 60

[11]

2017

Ultrasonic

anemometer

Not given Larger

than

200 mm

[26]

2018

Not given Ultrasonic

anemometer

840 mW Not given

This

work

Accelerometer-

based

anemometer

3.42 mW 50 mm in

diameter

3.4 Summary

In this chapter, we have proposed a new compact anemometer with ultra-low energy

consumption feature which could be suitable for energy constrained IoT environmental

sensing applications. The diameter of the sphere anemometer is 50 mm, and the

anemometer only consumes 3.42 mW when it is in operation. The proposed wind sensor

can be easily deployed in different places due to its compact size. The main contribution

of this work is that the proposed anemometer has low energy consumption feature

which allows it to be powered by using battery only. The wind force model and the 3D

dynamic model have been established to derive the equation of dynamics of the

anemometer, and the expressions between the wind velocity and the acceleration have

been given. A Gaussian-weighted moving average filtering algorithm is used to remove

the measuring noise, and we have demonstrated that the anemometer can measure both

wind speeds and wind directions accurately.


P a g e | 61

References

[1] Y. Zhu, B. Chen, M. Qin, and Q.-A. Huang, "2-D micromachined thermal wind

sensors—A review," IEEE Internet of Things Journal, vol. 1, no. 3, pp. 216-232,

2014.

[2] S. N. Daskalakis, G. Goussetis, S. D. Assimonis, M. M. Tentzeris, and A.

Georgiadis, "A uW backscatter-morse-leaf sensor for low-power agricultural

wireless sensor networks," IEEE Sensors Journal, vol. 18, no. 19, pp. 7889-

7898, 2018.

[3] N. Kumar and D. P. Vidyarthi, "A green routing algorithm for IoT-enabled

software defined wireless sensor network," IEEE Sensors Journal, vol. 18, no.

22, pp. 9449-9460, 2018.

[4] S. Mathew, Wind energy: fundamentals, resource analysis and economics.

Springer, 2006.

[5] S. Pindado, J. Cubas, and F. Sorribes-Palmer, "The cup anemometer, a

fundamental meteorological instrument for the wind energy industry. Research

at the IDR/UPM Institute," Sensors, vol. 14, no. 11, pp. 21418-21452, 2014.

[6] S. Pindado, J. Cubas, and F. Sorribes-Palmer, "On the harmonic analysis of cup

anemometer rotation speed: A principle to monitor performance and

maintenance status of rotating meteorological sensors," Measurement, vol. 73,

pp. 401-418, 2015.

[7] S. Pindado, A. Barrero-Gil, and A. Sanz, "Cup anemometers’ Loss of

performance due to ageing processes, and its effect on Annual Energy

Production (AEP) estimates," Energies, vol. 5, no. 5, pp. 1664-1685, 2012.


P a g e | 62

[8] M. Piotto, M. Dei, G. Pennelli, and P. Bruschi, "A miniaturized 2D solid state

anemometer based on thermal flow sensors," Procedia Chemistry, vol. 1, no. 1,

pp. 1463-1466, 2009.

[9] P. Bruschi, M. Dei, and M. Piotto, "A low-power 2-D wind sensor based on

integrated flow meters," IEEE Sensors Journal, vol. 9, no. 12, pp. 1688-1696,

2009.

[10] A. Wieser, F. Fiedler, and U. Corsmeier, "The influence of the sensor design on

wind measurements with sonic anemometer systems," Journal of atmospheric

and oceanic technology, vol. 18, no. 10, pp. 1585-1608, 2001.

[11] G. M. G. Lopes et al., "Development of 3-D ultrasonic anemometer with

nonorthogonal geometry for the determination of high-intensity winds," IEEE

Transactions on Instrumentation and Measurement, vol. 66, no. 11, pp. 2836-

2844, 2017.

[12] G. Bucci, F. Ciancetta, E. Fiorucci, D. Gallo, C. Landi, and M. Luiso, "A low-

cost ultrasonic wind speed and direction measurement system," 2013 IEEE

International Instrumentation and Measurement Technology Conference

(I2MTC), pp. 505-510, 20

[13] B. Allotta, L. Pugi, T. Massai, E. Boni, F. Guidi, and M. Montagni, "Design and

calibration of an innovative ultrasonic, arduino based anemometer," 2017 IEEE

International Conference on Environment and Electrical Engineering and 2017

IEEE Industrial and Commercial Power Systems Europe (EEEIC/I&CPS

Europe), pp. 1-6, 2017.

[14] Y. Zhu, B. Chen, M. Qin and Q. Huang, "2-D Micromachined Thermal Wind

Sensors—A Review," in IEEE Internet of Things Journal, vol. 1, no. 3, pp. 216-

232, June 2014.


P a g e | 63

[15] M.Laghrouchea, A.Adaneb, J.Bousseyc, S.Ameura, D.Meunierc, S.Tardud, "A

miniature silicon hot wire sensor for automatic wind speed measurements."

Renewable Energy 30.12(2005):1881-1896.

[16] B. Idjeri, M. Laghrouche, and J. Boussey, "Wind measurement based on MEMS

micro-anemometer with high accuracy using ANN technique," IEEE Sensors

Journal, vol. 17, no. 13, pp. 4181-4188, 2017.

[17] H.-B. Liu, N. Lin, S.-S. Pan, J. Miao, and L. K. Norford, "High sensitivity,

miniature, full 2-D anemometer based on MEMS hot-film sensors," IEEE

Sensors Journal, vol. 13, no. 5, pp. 1914-1920, 2013.

[18] S. Kim, T. Nam, and S. Park, "Measurement of flow direction and velocity using

a micromachined flow sensor," Sensors and Actuators A: Physical, vol. 114, no.

2-3, pp. 312-318, 2004.

[19] Z. Liu, F. Wang, Y. Zhang, Z. Jing, and W. Peng, "Low-Power-Consumption

Fiber-Optic Anemometer Based on Long-Period Grating With SWCNT

Coating," IEEE Sensors Journal, vol. 19, no. 7, pp. 2592-2597, 2019.

[20] H, Julian, "Wind speed measurements in an offshore wind farm by remote

sensing. A comparison of radar satellite TerraSAR-X and ground-based LiDAR

systems." (2015).

[21] C, Swift, "Passive Microwave Remote Sensing of Ocean Surface Wind Speed.

Surface Waves and Fluxes". Springer Netherlands, 1990.

[22] M. A. Mohandes and S. Rehman, "Wind Speed Extrapolation Using Machine

Learning Methods and LiDAR Measurements," in IEEE Access, vol. 6, pp.

77634-77642, 2018.


P a g e | 64

[23] R. Frehlich and N. Kelley, "Measurements of Wind and Turbulence Profiles

With Scanning Doppler Lidar for Wind Energy Applications," in IEEE Journal

of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 1,

no. 1, pp. 42-47, March 2008.

[24] B. Sun, W. Zhou, M.-Z. Yuan, E.-Q. Fang, and N. Gan, "A Cylindrical Vehicle-

Mounted Anemometer Based on 12 Pressure Sensors—Principle, Prototype

Design, and Validation," IEEE Sensors Journal, vol. 18, no. 17, pp. 6954-6961,

2018.

[25] G.-Z. Li, S. Zhao, and R. Zhu, "Wearable anemometer with multi-sensing of

wind absolute orientation, wind speed, attitude, and heading," IEEE Sensors

Journal, vol. 19, no. 1, pp. 297-303, 2018.

[26] H. B. Fan, J. L. Liu, and G. Z. Sun, "Design of Low-power Ultrasonic

Anemometer Based on STM32L476," IOP Conference Series: Materials

Science and Engineering, vol. 408, no. 1, p. 012036, 2018.

Chapter 5: Conclusions and Future Work

P a g e | 65

Chapter 4 An Environmental Sensing Platform

with Self-powered Standalone Weather Stations

Chapter 3 has proposed a novel and compact wind sensor with ultra-low energy

consumption feature which can be integrated into a weather station. In this chapter, we

will develop a weather station which can act as a wireless sensor node. The weather

station should be small in size and consume low energy to meet the requirements of IoT

applications. A solar energy harvesting system will be integrated into the sensor node

to make it to be self-powered. The weather station should have a stable wireless

connection, and different weather stations can be deployed in different places and

connected to build an environmental sensing platform.

4.1 Introduction

Real-time environmental data (e.g., temperature, humidity, lightness, wind speeds, and

wind directions) is needed in different applications such as weather monitoring/

forecasting, smart city, smart agriculture, air quality monitoring, and load prediction [1,

2]. Climate changes have received much attention recently, which makes it an active

and hot research area. To statistically study the climate change and the impacts of the

human activities on nature environment, we need to collect data like rainfall,

temperature, CO2, wind speed, and air quality from various places over the time [3].

Environmental monitoring can also be used in other applications. For example,

environmental sensors should provide accurate and real-time metrological data to

predict the load and the amount of energy generated from solar farm and wind turbine

to meet the requirements in the smart grid [4-6]. Additionally, to realize smart

agriculture, different sensors should be deployed over a wide region to monitor the crop


P a g e | 66

and environmental conditions in the farm [7-9].

Different sensors are widely deployed to realize environmental monitoring applications,

and how to connect those sensors to allow real-time data transmission is a challenging

problem. With the rapid growth of IoT technologies, some machine type

communication (MTC) protocols like NB-IoT, LoRaWAN and Sigfox have emerged

and gained wide attention from many researchers [10]. They have low energy

consumption and wide coverage, which is suitable for IoT applications. Among them,

LoRaWAN allows to set up a private network, which can be integrated into different

global network platforms (e.g., The Things Network) [11]. Thus, LoRaWAN has been

selected for this project.

The power consumption of the sensor node is another main design challenge, which is

important to determine the lifespan of the node and maintain the overall performance

of the whole sensor network [12]. However, few research has been focused on the

energy consumption of the sensor node [13, 14]. It will increase the cost and difficulty

of the deployment of the sensing system if power lines are used to power the end devices.

The sensor node could be designed to be powered by the battery to increase deployment

flexibility. Energy-saving techniques and energy harvesting methods should also be

implemented to reduce battery replacement frequency and save the maintenance cost.

Solar energy has a relatively high power density, and solar panel is used to provide

energy to the sensing device.

This chapter proposes a new design of smart environmental sensing system with self-

powered sensor nodes, LoRaWAN communication network, and real-time data

accessible cloud. The wireless sensor node is embedded with a sleep mode control to

save operating power. Moreover, a solar energy harvesting powering method is

developed for the sensor. By integrating the solar energy harvesting method, the sensor

node can become self-powered without the need of changing the battery. LoRaWAN is


P a g e | 67

used to provide a real-time, robust, and wide range connectivity for the sensor nodes.

The real-time data gathered from the sensor nodes can be easily monitored and obtained

from a Could App (Cayenne). The total cost of one sensor node is around 100 pounds.

4.2 Hardware and Software

4.2.1 Hardware

Wireless Weather Stations

The weather station can be viewed as a sensor node composed of four parts which are

the Power Management Unit (PMU), the sensing unit, the microcontroller unit (MCU),

and the communication unit as shown in Fig. 4-1. The sensors can measure different

environmental parameters, and the measured data can be read by the MCU. The MCU

will process the raw data and transfer the data obtained from the sensors into a single

payload which is ready to be sent to the cloud. Then, the LoRaWAN communication

unit can transmit the payloads containing the value obtained from the environmental

sensors to the LoRaWAN gateways.

Fig.4- 1 The architecture of the weather station.


P a g e | 68

Table 4- 1 The characteristics of sensors

Model Figure Functions Size Power

consumption

BME280

Temperature,

humidity,

air pressure

40 mm *

20 mm

< 13.32μW

MP503

Carbon

monoxide,

alcohol,

acetone

40 mm *

20 mm

< 300 mW

LM386

Loudness 20 mm *

20 mm

< 25 mW

GL5528

brightness 20 mm *

20 mm

< 15 mW


P a g e | 69

Table 4- 2 The Specifications of Parts of the Sensor Node

Name Model Function

Microcontroller Unit

(MCU)

Sodaq Mbili Data processing and power

management

LoRaWAN

communication module

RN2483 Data communication

6000 mAh LiPo battery KC 906090P Providing power to the device

3 W solar panel SKU 313070001 Providing power to the

battery

The main feature of the sensor node is that the whole device can be self-powered so

that the battery does not need to be replaced manually. To realize this function, a 3W

solar panel and a 6000 mAh Lithium Polymer (LiPo) battery are used to provide energy

to the whole sensor node. The PMU of the sensor node has two Japanese Solderless

Terminal (JST) ports to connect to the solar panel and the LiPo battery. Since the

harvested energy from the solar is not stable and consistent, and the battery is used to

provide a steady 3.7 V output voltage to ensure that each module of the sensor node can

work properly. The solar panel with a maximum output voltage of 6 V is used to charge

the LiPo battery under the control of the PMU. Four different sensors which are the

brightness sensor, loudness sensor, the air quality sensor, and the TPH sensor which can

measure the temperature, air pressure, and humidity are used in each sensor node. More

information about the four sensors can be found in Table 4-1. All the sensors are

connected to the Sodaq Mbili board by using grove connections, which can supply

energy to the sensors and allow the communication between the sensors and the MCU.

The Sodaq Mbili board has ten grove connectors, and more sensors can be connected

to the board depends on the applications.


P a g e | 70

The Sodaq Mbili uses a Microchip ATmega 1284p as its microcontroller which has

128kB ISP flash memory with read-while-write capabilities, 16kB SRAM and a real-

time counter. The MCU is designed to support low energy applications, and the current

consumption is 0.4 mA in the active mode and 0.6 μA in the power saving mode when

it operates at 1MHz, 1.8 v, 25 °C. The real-time counter embedded in the MCU can be

used to put the device into the sleep mode or wake up the board from the sleep mode at

specific times which could further reduce the energy consumption of the sensor node.

Another main function of the MCU is that it can read the data gathered by the sensors,

and the measured results are encoded to the payloads. The payloads are transmitted to

the LoRaWAN gateway by using the RN2483 LoRaWAN transceiver, as shown in Fig.

4-2. The instant transmission current consumption of the RN2483 is 44.5 mA, and the

idle current consumption is 3.1 mA when the transceiver is connected to a 3.6V power

supply. One of the operating frequency bands of the transceiver is 863 MHz to 870

MHz, which supports the LoRaWAN operating frequency in Europe.

Fig.4- 2 RN2483 LoRaWAN communication module.


P a g e | 71

The wireless sensor node is held in an IP66 dustproof and waterproof box in congestion

with a 3D printed base, as shown below. The box has a transparent cover which is made

of polycarbonates (PC) and an opaque base which is made of acrylonitrile butadiene

styrene (ABS). The 3D printed black base is placed at the bottom of the sensor box,

which is used to mount sensors, the battery, and the Sodaq Mbili board. The sensor box

is sealed to an IP66 standard to prevent the rain. The inside look and the outside look

of the sensor node are presented in Fig. 4-3 and Fig. 4-4.

Fig.4- 3 The inside look of the sensor node.


P a g e | 72

Fig.4- 4 The outside look of the sensor node.

Fig.4- 5 The LoRaWAN gateway.


P a g e | 73

LoRaWAN Gateway

The wireless environmental sensing nodes will be connected to the gateway by using

LoRaWAN protocol to form a wireless sensor network. To set up the LoRaWAN based

environmental sensing network, an 8 channels LoRaWAN gateway is used to provide

the LoRaWAN coverage to the sensor nodes. The indoor LoRaWAN gateway is built

based on the Raspberry Pi 3 and the RAK831 concentrator module (see Fig. 4-5). The

RAK831 is a multi-channel high-performance transmitter/receiver module designed to

receive up to 8 LoRa packets with different spreading factors at the same time on

multiple channels. It operates in the 863 MHz to 870 MHz frequency band and supports

both the LoRa and Frequency Shift Key (FSK) modulation techniques. A Raspberry Pi

3 acts as a host board to control the RAK831 frontend. A 5V and 2A power supply needs

to connect to the Raspberry Pi 3 to activate both the Raspberry Pi 3 and the RAK831.

4.2.2 Software

The wireless sensor node is developed based on the Sodaq Mbili which is based on the

Arduino development board. To program the sensor node and LoRaWAN transceiver,

we write all the programs by using the open-source Arduino Software (IDE). The

Arduino IDE is written in the programming language Java, and it supports C and C++.

The code for the node plays an essential role to guarantee the environmental sensing

platform can work properly. It needs to put the MCU and the sensors into sleep mode

to save energy and wake up them by using the interference generated by the real-time

clock to collect environmental data. The working diagram of the software is displayed

in Fig. 4-6. The activation method of the LoRaWAN connection, as well as activation

keys such as Network Session Key (NwkSKey), Application Session Key (AppSKey),

and Application Key (AppKey), should also be initialized in the code. Various sensed

environmental parameters such as temperature, humidity, loudness, and air pressure

need to be encoded to payloads, which can be interpreted by the Cayenne cloud.


P a g e | 74

Fig.4- 6 Software working diagram

The gateway will upload the received message from sensor nodes to a cloud platform,

namely The Things Network (TTN). To allow the received data to be sent to The Things

Network server, the Raspberry Pi 3 should be connected to the Internet by using 3G/4G,

Wi-Fi, or Ethernet. The working diagram of the connections between the sensor nodes,

gateway and, the cloud is displayed in Fig. 4-7. The gateway should be registered to

TTN to transmit the received data from the sensor nodes to the TTN console. Firstly,

the gateway needs to be registered in the TTN platform, and the TTN will automatically

generate the gateway ID and the gateway keys. The gateway ID and the gateway key

should be manually entered to the host board. Then, the status of the gateway in TTN

will be changed to connected, and the gateway is registered successfully in TTN console.


P a g e | 75

Fig.4- 7 The architecture of the environmental sensing platform.

The Things Network and MyDevice Cayenne

TTN is a free global open LoRaWAN network provider established in the Netherlands,

and there are more than 6000 LoRaWAN gateways registered in TTN all over the world.

To connect LoRaWAN end device to the TTN platform, a registered gateway in the

TTN is necessary. Once a LoRaWAN gateway is registered in the TTN, it provides

LoRa coverage to nearby sensor nodes. For a LoRaWAN end device, it also needs to be

registered to the TTN first before sending and receiving payloads. After that, a nearby

TTN LoRaWAN gateway can extract the payload sent by the node and, the payload can

be received and presented in the user’s TTN account. However, the data presentation

and storage function of the TTN is limited, and therefore in this proposed environmental

sensing platform, a data storage and visualization cloud named myDevice Cayenne is

integrated to the TTN to present and store the measured data. The payload sent by the

end device should be encoded by using Cayenne Low Power Payload (Cayenne LPP)

and, the Cayenne cloud can then decode the payloads correctly. The received payload

in TTN can be synchronously transmitted to myDevice Cayenne by using Application

Programming Interface (API) to realize real-time data communication. Finally, all


P a g e | 76

environmental data measured by sensors can then be visualized and stored in the

Cayenne. The users can access the data from the Cayenne IoT website or mobile APP

in real time.

4.3 Measurement and Discussion

4.3.1 Deployment

Three wireless sensor nodes have been deployed across the University of Liverpool as

a pilot trial for more than three months sensing up to 5 environmental parameters

including temperature, air pressure, humidity, loudness, and air quality. They are placed

in both indoor and outdoor conditions, as displayed in Fig. 4-8. Since we aim to make

each sensor node be a standalone device which can be powered by using solar energy,

the device is placed near the window in indoor circumstances to obtain enough sunlight.

The LoRaWAN gateway is an indoor gateway, and it is placed in a typical office, as

shown in Fig. 4-8. The gateway is placed near the window facing the city center to

cover more end users.


P a g e | 77

Fig.4- 8 The deployment of sensor nodes and the LoRaWAN gateway. (a) The

sensor node in indoor condition; (b) The sensor node in indoor condition; (c) The

sensor nodes in outdoor condition; (d) The LoRaWAN gateway in indoor condition.

4.3.2 Measurement Results

Fig.4- 9 The measured temperature over 5 days.


P a g e | 78

Fig.4- 10 The measured air pressure over 5 days.

Fig.4- 11 The measured humidity over 5 days.

Fig.4- 12 The measured loudness over 5 days.


P a g e | 79

Fig.4- 13 The measured air quality over 5 days.

All the measured environmental data will be stored and presented in the myDevice

Cayenne cloud. Fig. 4-9 to Fig. 4-13 present 5-days continuous environmental

monitoring results of 5 measuring parameters from one device. Fig. 4-9 presents the

variation of temperature over 5 days, and it can be easily found that there is a peak

during the daytime and a trough at night every day. It agrees with the common

knowledge of the temperature change during a day. The air pressure and humidity are

displayed in Fig. 4-10 and Fig. 4-11, the variation of these two parameters is relatively

small according to the measured results. The loudness level shown in Fig. 4-12 has a

similar trend as the temperature variation. It is because more activities will be carried

out during daytime which could make noise. Also, the noise level has sudden changes,

which also agrees with the nature of the noise generated by different activities. VOC

measures the concentration of volatile organic compounds in the air and can be used to

describe the air quality. VOC varies in a small region, and it does not have significant

differences between day and night.


P a g e | 80

Fig.4- 14 The measured results of three sensor nodes.

A small wireless sensor network consisting of three sensor nodes which is used to

monitor the ambient environment has been successfully built. Fig. 4-14 is an overview

of all the environmental data obtained from three wireless sensor nodes. All these

parameters are received and displayed in real time. The RSSI value can be used to tell

the received signal strength of each sensor node to guarantee the connection is stable.

The battery level indicates the voltage of the supply battery, and it can be used to

monitor the remaining energy in the battery. It is also a critical parameter to investigate

if the sensor node does not work properly. The sensor nodes can be deployed over a

wide region which is covered by the LoRaWAN gateway. The users can have access to

all the measured results from all three sensors deployed in different places to monitor

and manage corresponding activities.


P a g e | 81

4.3.3 Coverage Analysis

Fig.4- 15 The coverage test of LoRaWAN.

To do the coverage test, the team placed the gateway on the first floor of the Electrical

Engineering building, which is indicated by the orange icon. The position labelled by

using the green icon can receive the LoRaWAN signal, and there is no signal at the

place marked by the red icon. The two places with no signal are due to the obstruction

of the tall buildings. Based on the test, it can be found that the LoRaWAN can achieve

better coverage with line-of-sight, and the obstruction can block the signal. Therefore,

the deployed position of the gateway should be carefully chosen to maximise the

coverage, and it will be better to put the gateway in a high place with less obstruction.


P a g e | 82

Fig.4- 16 The coverage distance of the LoRaWAN gateway.

The longest coverage distance obtained from the previous test is 366.63 m, as shown in

Fig. 4-16, which is smaller than the distance stated in the datasheet of the LoRaWAN.

The measurements agreed with some of the experiments did by other researchers.

Junqing measured a coverage of 500 m, and Erbati obtained a coverage of 300 m from

an outdoor gateway [15, 16]. There are few factors to cause the limited coverage of the

LoRaWAN gateway. The first is that it is an indoor gateway, and the transmission power

of the indoor gateway is relatively low. Since it is placed in the indoor environment,

and the building will obstruct the signal, which can reduce the coverage as well. The

penetration loss is 20 dBm for indoor devices [17]. Also, the gateway is placed on the

first floor, and wider coverage can be achieved if the gateway is placed in a high place

[18]. Another cause of the short coverage is because the gateway is placed in the urban

area. Different obstacles, such as high rise building and tall street furniture will cause

none line-of-sight (LoS) propagation and the multipath effect, which will reduce

coverage area as well[19].


P a g e | 83

To statistically analyse the coverage distance of the proposed LoRaWAN environmental

sensing system, Okumura Hata model is used to theoretically study the fade margin and

to determine whether the theoretical value is consistent with the measured value. The

calculated path loss by using the Okumura Hata model is 108 dBm. According to the

specification of the RN2483 LoRaWAN transmitting module, the output power could

be 10.4 dBm. Since, the gateway is placed in the indoor condition, another 20 dBm path

loss should be added to the total path loss. The received signal strength could then be

calculated to be -118.6 dBm, which is very closed to the LoRaWAN lowest sensitivity

at the SF 7. It could be proved that the measured results agree well with the theorical

values.

4.3.4 Energy Performance Analysis

The energy consumptions of some sensors are very low, and a TPH sensor only

consumes 13.32 μW. However, the air quality sensor has an energy consumption of 300

mW when it is in operation, and it will drain the battery very quickly if it is continuously

switched on. To reduce the energy consumption of the whole device, we add a sleeping

function to the MCU. The whole device will be put into the sleep mode with extremely

low energy consumption. The device will only be activated in a short period, and it will

be sleep again after the sensing and transmission of environmental parameters. Since

the MCU and sensors will be in the sleep mode most of the time, and sensors with high

energy consumption will not have huge impacts on the overall energy performance of

the device.

To numerically study the energy consumption of the sensor node, we measured the

current consumption of the sensor node over a period, and the results are shown in Fig.

4-17. The total wake-up time of the sensor node is 4.5 s, and there are five current peaks

which represent the board initialization, sensor initialization, transmission, and two


P a g e | 84

receiving windows. The maximum current consumption is during the LoRaWAN

transmission, which is about 52 mA. However, it only lasts for 0.067 s, and only 12.395

mJ will be consumed during each transmission. The sensor node will go to the sleeping

mode by the end of the second receiving window, and the current consumption during

the sleep mode is only 3.7 mA which makes the overall power consumption of the

sensor node to be 13.69 mW.

Fig.4- 17 The current consumption of the sensor node.

Fig.4- 18 The change of the voltage level in 3 months.


P a g e | 85

Fig.4- 19 The change of the voltage level in 1 month.

We continuously monitored the energy remained in the battery of the weather station

over three months starting from December 17th, 2018. In this test, we use the output

voltage level of the battery to determine the remaining energy in the battery. The battery

voltage level over three months can be found in Fig.4-18. Fig.4-19 presents a detailed

voltage level change over 30 days starting from January 5th, 2019. Form Fig. 4-18, we

can find that the voltage level drops and goes up on different days. It means that the

sensor node consumes the remaining energy in the battery when the solar panels cannot

extract enough energy. The main factor which could affect the amount of harvested

energy from solar panel is the weather condition. The voltage level continuously drops

due to the consistent rainy weather from December 17th, 2018 to January 22th, 2019.

However, the solar panel will charge the battery if the weather conditions are good, and

after 10 days, the voltage level goes from 4.18 V to 4.20 V (see Fig. 4-19) which means

that more energy is received and charged to the battery than the amount of the energy

that the sensor node consumes. It demonstrates that more energy is harvested by using

the solar panels than the energy consumed by the sensor node, and the device can then

be a self-powered sensor node without changing the battery.

The deep charging cycle of the KC 906090P battery used in the sensor node is more

than 1000 times. Solar panel is used to provide power to the battery, and the battery can

be quickly recharged in a relatively short time. This charging method could be defined


P a g e | 86

as the shallow discharge. Shallow discharge is better to battery compared with deep

discharge, which means the battery can be used longer when using the solar panel to

power it. The battery can be used to power the sensor node for 2 months in one full

charge - recharge cycle, and the battery life could therefore be more than 166 years

theoretically.

4.3.5 Discussions

Table 4- 3 Comparison of the proposed weather station with other designs

Ref

(year)

Platform Environmental

parameters

Energy

consump

tion

Cloud Energy

harvesting

Cost

[20]

2015

Robin

Z530L

Temperature, humidity,

air pressure, wind speed

Not

given

Yes No Not

given

[21]

2017

STM32F1

03VETb


light intensity, wind

speed, wind direction

68.75

mW

No Solar panels

and wind

Not

given

[22]

2017

Not given Temperature, humidity,

wind speed, rain, solar

radiation

Not

given

Yes No Not

given

[23]

2017

Raspberry

PI and

Arduino


air pressure, wind speed,

wind direction

Not

given

Yes No GBP

127.5

[24]

2018

Raspberry

PI


air pressure, wind speed,

wind direction,

2.775 W Yes Solar panels GBP

231.81

[25]

2019

Arduino Temperature, humidity,

air pressure

Not

given

No Solar panel Not

given


P a g e | 87

[26]

2018


air pressure, rain

detection

Not

given

Yes No Not

given

This

work


air pressure, brightness,

loudness, air quality

13.69

mW

Yes Solar panel GBP

112.59

A comprehensive comparison between different weather stations has been listed in the

above table. It can be concluded that all the weather stations can sense some basic

environmental parameters such as temperature, humidity, and air pressure. However,

there are some advanced functions such as the connection to the cloud and the energy

harvesting which are not included in all the weather stations. Only one of the seven

weather stations include both the cloud connection and the energy harvesting features.

Also, most of the researches did not investigate the energy consumption of the devices,

and the weather station including both cloud connection and energy harvesting function

consumes 2.775 W which is extremely high for an electronic device, and it may not be

able to be powered by only using the battery [2]. In this work, we proposed a low cost

and low energy consumption weather station which only consumes 13.69 mW, and it

can be integrated to the TTN cloud and the Cayenne cloud so that the users can have

access to the measured data in real time. We have demonstrated that the weather station

can be self-powered by using the energy received from solar panels.

An environmental sensing platform has been successfully built by integrating three

weather stations and two clouds. The environmental parameters collected from different

weather stations can be transmitted to the cloud by using LoRaWAN, and the measured

data can be presented and stored in the cloud to be easily accessed by the users. In this

sensing platform, each sensor node has a compact size and can be self-powered so that

the sensor nodes can be flexibly deployed. The sensing platform can be used to collect

various environmental parameters in different applications to achieve monitoring and


P a g e | 88

management purposes.

4.4 Summary

The environmental sensing platform consists of three wireless weather stations has been

proposed. We have demonstrated the excellent performance of the sensing platform

from three aspects, which are the real-time data presentation and storage in the Cayenne

cloud, the wide coverage, and the integration of self-powered weather stations. Since

we aim to densely deploy weather stations which could cover a wide area, the wireless

connection and the energy consumption of the sensor nodes are key design challenges

need to be overcome. In this design, we use the LoRaWAN communication method to

provide a wide wireless connection for the sensor nodes. The coverage of the

LoRaWAN gateway can be 366.63 m in the urban area. We reduce the energy

consumption of the weather station to 13.69 mW and use a 3 W solar panel to power

the device. Based on the tests, it can be proved that the solar panel can provide enough

energy to make the device to be self-powered.

4.5 Reference:



station for small-scale wind farm site selection," pp. 1-5, 2018.


"Renewable Powered Portable Weather Update Station," pp. 374-377, 2019.

[3] G. Mois, S. Folea, and T. Sanislav, "Analysis of three IoT-based wireless sensors


P a g e | 89

for environmental monitoring," IEEE Transactions on Instrumentation and

Measurement, vol. 66, no. 8, pp. 2056-2064, 2017.

[4] P. Dehghanian, B. Zhang, T. Dokic, and M. Kezunovic, "Predictive risk

analytics for weather-resilient operation of electric power systems," IEEE

Transactions on Sustainable Energy, vol. 10, no. 1, pp. 3-15, 2019.

[5] H. Sangrody, M. Sarailoo, N. Zhou, N. Tran, M. Motalleb, and E. Foruzan,

"Weather forecasting error in solar energy forecasting," IET Renewable Power

Generation, vol. 11, no. 10, pp. 1274-1280, 2017.




[7] J. Liu, Y. Chai, Y. Xiang, X. Zhang, S. Gou, and Y. Liu, "Clean energy

consumption of power systems towards smart agriculture: roadmap, bottlenecks

and technologies," CSEE Journal of Power and Energy Systems, vol. 4, no. 3,

pp. 273-28, 2018.

[8] N. Ahmed, D. De, and I. Hussain, "Internet of Things (IoT) for Smart Precision

Agriculture and Farming in Rural Areas," IEEE Internet of Things Journal, vol.

5, no. 6, pp. 4890-4899, 2018.

[9] N. Gondchawar and R. S. Kawitkar, "IoT based smart agriculture," International

Journal of advanced research in Computer and Communication Engineering,

vol. 5, no. 6, pp. 2278-1021, 2016.

[10] X. Zhang, M. Zhang, F. Meng, Y. Qiao, S. Xu and S. Hour, "A Low-Power

Wide-Area Network Information Monitoring System by Combining NB-IoT

and LoRa," in IEEE Internet of Things Journal, vol. 6, no. 1, pp. 590-598, Feb.


P a g e | 90

2019.

[11] J. Haxhibeqiri, E. De Poorter, I. Moerman, and J. Hoebeke, "A survey of

lorawan for iot: From technology to application," Sensors, vol. 18, no. 11, p.

3995, 2018.


[13] E. Kanagaraj, L. M. Kamarudin, A. Zakaria, R. Gunasagaran, and A. Y. M.

Shakaff, "Cloud-based remote environmental monitoring system with

distributed WSN weather stations," 2015 IEEE SENSORS, pp. 1-4, 2015.

[14] R. C. Brito, F. Favarim, G. Calin, and E. Todt, "Development of a low cost

weather station using free hardware and software," 2017 Latin American

Robotics Symposium (LARS) and 2017 Brazilian Symposium on Robotics

(SBR), pp. 1-6, 2017.

[15] M. M. Erbati, G. Schiele, and G. Batke, "Analysis of LoRaWAN technology in

an Outdoor and an Indoor Scenario in Duisburg-Germany," 2018 3rd

International Conference on Computer and Communication Systems (ICCCS),

pp. 273-277, 2018.

[16] J. Zhang, A. Marshall, and L. Hanzo, "Channel-Envelope Differencing

Eliminates Secret Key Correlation: LoRa-Based Key Generation in Low Power

Wide Area Networks," IEEE Transactions on Vehicular Technology, vol. 67, no.

12, pp. 12462-12466, 2018.

[17] B. Vejlgaard, M. Lauridsen, H. Nguyen, I. Z. Kovács, P. Mogensen, and M.

Sorensen, "Interference impact on coverage and capacity for low power wide

area IoT networks," 2017 IEEE Wireless Communications and Networking

Conference (WCNC), pp. 1-6, 2017.


P a g e | 91

[18] A. V. T. Bardram, M. D. Larsen, K. M. Malarski, M. N. Petersen, and S. Ruepp,

"LoRaWan capacity simulation and field test in a harbour environment," 2018

Third International Conference on Fog and Mobile Edge Computing (FMEC),

pp. 193-198, 2018.

[19] A. Ikpehai et al., "Low-Power Wide Area Network Technologies for Internet-

of-Things: A Comparative Review," IEEE Internet of Things Journal, 2018.

[20] E. Kanagaraj, L. M. Kamarudin, A. Zakaria, R. Gunasagaran, and A. Y. M.

Shakaff, "Cloud-based remote environmental monitoring system with

distributed WSN weather stations," 2015 IEEE SENSORS, pp. 1-4, 2015.




[22] A. Munandar, H. Fakhrurroja, M. I. Rizqyawan, R. P. Pratama, J. W. Wibowo,

and I. A. F. Anto, "Design of real-time weather monitoring system based on

mobile application using automatic weather station," 2017 2nd International

Conference on Automation, Cognitive Science, Optics, Micro Electro-

Mechanical System, and Information Technology (ICACOMIT), pp. 44-47, 2017.

[23] R. C. Brito, F. Favarim, G. Calin, and E. Todt, "Development of a low cost

weather station using free hardware and software," 2017 Latin American

Robotics Symposium (LARS) and 2017 Brazilian Symposium on Robotics (SBR),

pp. 1-6, 2017.



station for small-scale wind farm site selection," 2018 IEEE International

Instrumentation and Measurement Technology Conference (I2MTC), pp. 1-5.


P a g e | 92

IEEE, 2018.


"Renewable Powered Portable Weather Update Station," 2019 International

Conference on Robotics, Electrical and Signal Processing Techniques

(ICREST), pp. 374-377, 2019.

[26] M. Kusriyanto and A. A. Putra, "Weather Station Design Using IoT Platform

Based On Arduino Mega," 2018 International Symposium on Electronics and

Smart Devices (ISESD), pp. 1-4, 2018


P a g e | 93

Chapter 5 Conclusions and Future Work

5.1 Summary

In this thesis, an accelerometer is used to detect wind speeds and wind directions. The

power consumption of the anemometer is only 3.42 mW. We have also applied some

latest IoT technologies to build an energy-saving anemometer and a modern

environmental sensing platform. LoRaWAN, as one of the state-of-art Low Power Wide

Area Network (LPWAN) communication methods has been used to transmit messages

between the sensor node and the gateway to achieve low energy consumption in the

sensor node and wide coverage. To increase the flexibility of the deployment of the

sensor nodes, we make the device to be self-powered by adopting energy saving

techniques and using a solar panel to charge the device. The energy consumption of the

proposed sensor node is only 13.69 mW, which is significantly smaller than other

weather stations. The environmental sensing platform integrates with two clouds which

are the The Things Network (TTN) and the Cayenne to achieve real-time data

presentation and data storage.

In chapter 2, we have surveyed two types of communication protocols, namely short-

range communication protocols and LPWAN protocols. It could be concluded that

LPWAN protocols have the merits of wide coverage and low energy consumption,

which is more suitable to transmit data in an energy-constrained wireless sensor

network. Three LPWAN communication methods which are LoRaWAN, SigFox, and

NB-IoT are discussed, and the LoRaWAN has the advantages of flexible private

deployment and low energy consumption feature which is more suitable to form a

private network. A brief introduction of the wireless sensor network is also discussed

in this chapter to provide basic knowledge about the architecture of the wireless sensor

network.


P a g e | 94

A compact anemometer with ultra-low energy consumption has been proposed in

chapter 3. The anemometer is composed of a sphere 3D printed sensor case and a 6-

axis accelerometer. The 3D dynamic model of the anemometer has been built, and the

equations of the dynamics of the wind sensor have been derived as well. The diameter

of the sphere anemometer is 50 mm, and the anemometer only consumes 3.42 mW

when it is in operation. The wind sensor can be easily deployed in different places and

can be powered by using batteries only, which could increase the flexibility of

deployment. A Gaussian-weighted moving average filtering algorithm is used to

remove the measuring noise, and we have demonstrated that the anemometer can

measure both the wind speeds and wind directions accurately.

In chapter 4, we have presented an environmental sensing platform, including both the

hardware and the software. The environmental sensing platform is consisted of three

wireless weather stations, and each of them is connected to the gateway to receive and

transmit data. In this design, we use the LoRaWAN communication method to provide

a wide wireless connection for the sensor nodes. The coverage of the LoRaWAN

gateway can be 366.63 m in the urban area. The information obtained from the weather

stations can be transmitted to the TTN and the Cayenne cloud where received data can

be presented and stored in real time. We have investigated the energy consumption of

the weather station, which is only 13.69 mW. It can be proved that the weather station

can be self-powered by harvesting solar energy based on the measurements over a

month.

5.2 Key Contributions and Limitations

This work has provided a thorough study on the environmental monitoring applications

from an individual wind sensor to the whole environmental monitoring system. The key

contributions and limitations are discussed as follows.


P a g e | 95

⚫ A Low Energy Consumption Anemometer

The most important contribution in this chapter is the development of an ultra-low

energy consumption anemometer. There are various ways to measure wind speeds

and directions. Some latest measuring methods like ultrasonic sensors and hot-film

sensors typically dissipate a large amount of power. A novel method to measure the

wind speeds and directions has been developed which are using a 6-axis

accelerometer to measure the changes of the accelerations induced by the wind

force and then to calculate the corresponding wind speed. The energy consumption

of the proposed anemometer is only 3.42 mW, which is significantly smaller than

other anemometers. However, the measurement accuracy of the anemometer

should be further improved. To increase the measurement accuracy of the

accelerometer-based anemometer, a Gaussian-weighted moving average filtering

algorithm is applied to remove the measuring noise.

⚫ The development of an environmental sensing platform

In this chapter, we have developed an environmental sensing platform including,

wireless sensor nodes, communication network, and the data visualization and

storage cloud. Most reported weather stations do not include cloud, and in this work,

we utilize two cloud servers to transmit, visualize, and store the measured data.

Also, many researchers do not investigate the energy performance of the weather

stations, which makes the energy consumptions of the designed weather stations to

be very high. We have implemented some energy saving techniques such as putting

the device into the sleep mode and using low energy consumption communication

module to reduce the overall energy consumption of the proposed device to be only

13.69 mW. We also integrate a 3 W solar panel to the weather station to make the

whole device to be self-powered. The performance of the real-field deployment of

the environment sensing platform including coverage and maximum data flow

needs to be further studied. More functions could be added to the sensing system


P a g e | 96

based on the needs of different applications. Also, the package design of the sensor

node needs to be improved to protect the sensing device in extreme weather

conditions.

5.3 Future Work

Based on the conclusions above and considering the limitations of the work existed,

future research could be carried out in the following areas.

⚫ The package design of sensors is a challenging topic. The size of the sensor node

should be designed to be compact to reduce the manufacturing cost and increase

deployment flexibility. Different modules, such as sensing modules,

communication modules, power management modules should be placed together

in a limited space. There are various deployment requirements on different modules.

For example, some environmental sensors like gas sensor and rain detection sensor

should be placed in contact with the outside environment. However, the batteries

should be placed in an enclosed environment to prevent damage from rain and short

circuits. The sensor box of the anemometer can be designed like a “golf ball” to

reduce the turbulence induced by the wind and improve the measuring accuracy

and sensitivity.

⚫ The proposed weather station can now measure 6 environmental parameters which

are the brightness, loudness, temperature, air pressure, humidity, and air quality. In

the future, more sensors with more functions should be added to the existed device

to meet more requirements from other applications. For example, to monitor the

structural health level, an accelerometer is used to measure the vibrations of

buildings and bridges. Also, the proposed anemometer can only transmit data to


P a g e | 97

the smartphone by using Bluetooth. The next step of the work can be integrating

the anemometer with the proposed weather station to measure the wind speeds and

wind direction。

⚫ Currently, we only collected basic environmental parameters for three months

around the campus. More data can be collected in different places such as city

centers, factories, and rural areas for a long-time span. The coverage of the

environmental system should be further tested to meet the requirement of IoT

applications. The collected data can provide sources to analyze the measured

environmental changes over the years across different places and study how human

activities can affect the environments. Currently, we only measure and present the

environmental parameters, and further studies can be carried out to analyze and

predict the condition of the ambient environment based on the collected results.

For example, some researchers found that the concentration of the CO2 level in a

room is related to the number of people inside the room, and then they use

measured CO2 level to predict the number of people in the room.

the iot based environmental sensing platform

Documents